A virus-like particle vaccine for epidemic Chikungunya virus protects nonhuman primates against infection

Abstract

Chikungunya virus (CHIKV) has infected millions of people in Africa, Europe and Asia since this alphavirus reemerged from Kenya in 2004. The severity of the disease and the spread of this epidemic virus present a serious public health threat in the absence of vaccines or antiviral therapies. Here, we describe a new vaccine that protects against CHIKV infection of nonhuman primates. We show that selective expression of viral structural proteins gives rise to virus-like particles (VLPs) in vitro that resemble replication-competent alphaviruses. Immunization with these VLPs elicited neutralizing antibodies against envelope proteins from alternative CHIKV strains. Monkeys immunized with VLPs produced high-titer neutralizing antibodies that protected against viremia after high-dose challenge. We transferred these antibodies into immunodeficient mice, where they protected against subsequent lethal CHIKV challenge, indicating a humoral mechanism of protection. Immunization with alphavirus VLP vaccines represents a strategy to contain the spread of CHIKV and related pathogenic viruses in humans.

Figure 1. Characterization of CHIKV E pseudotyped lentiviral vectors

(a) Schematic representation of the CHIKV genome and CHIKV E expression vector used for incorporation of CHIKV E from strains 37997 and LR2006 OPY-1 into pseudotyped lentiviral vectors. The CHIKV genome consists of nonstructural polyproteins nsP1, nsP2, nsP3 and nsP4 and structural polyproteins capsid (C) and envelope (E: E3, E2, 6K and E1) (top). The polypeptide E genes from strains 37997 and LR2006 OPY-1 were inserted into an expression vector (bottom; schematic representation of vector is in Supplementary Fig. 1a, top). (b) Infectivity of the indicated pseudotyped lentiviral vectors in several CHIKV-permissive cell lines including 293A human renal epithelial, HeLa cervical epithelial, Vero renal epithelial and baby hamster kidney (BHK) cells. The pseudotyped vectors were standardized by HIV-1 Gag p24 (left) or the indicated concentration of p24 and used to infect 293A cells (right). After incubation with pseudotyped vectors for 24 h, cells were lysed and luciferase activity was measured. (c) pH-dependent entry of CHIKV pseudotyped lentiviral vectors incubated in the presence of the indicated amounts of ammonium chloride (left) and chloroquine (right). Data are presented as the percentage of activity at the indicated dose relative to activity with no treatment. (d) Neutralization measured with pseudotyped lentiviral vectors in sera from immune mice infected with CHIKV strain S-27 (ATCC). Sera were incubated at the indicated dilutions with VSV-G, CHIKV strain 37997 or LR2006 OPY-1 E-pseudotyped lentiviral vectors and the mixture infected to 293A cells. No inhibition was observed with control non-immune antisera.

Figure 2. Characterization of CHIKV VLPs

(a) Schematic representation of CHIKV C-E or E expression vectors used for DNA vaccine and VLP production. The CHIKV structural polyproteins capsid plus envelope (C-E) or E alone from strains 37997 and LR2006 OPY-1 were inserted into a CMV/R expression vector (Methods). 293T cells were transfected with the indicated plasmids. Expression was measured 48 h after transfection by Western blotting as described (Supplementary Methods) (b) VLPs were purified (Methods)from the supernatants of 293F cells transfected with C-E expression vector (C-E₃₇₉₉₇) (left) 72 h after transfection. Each fraction was characterized for its buoyant density (left upper panel) and protein content (left lower panel) by Western blot analysis. The fractionated VLPs were observed by transmission electron microscopy with magnification 20,000× (right, bar 100 nm). (c) Cryo-EM reconstruction of CHIKV VLP reveals its structural similarity to alphaviruses. Shaded-surface representation of the 3D density map of CHIKV VLP is viewed along an icosahedral 2-fold axis and resembles Sindbis virus (Supplementary Fig. 2; see also refs. 8,9,18). The white triangle marks the boundary of an icosahedral asymmetric unit. The numbers show the positions of the icosahedral 2-, 3-, and 5-fold axes limiting an asymmetric unit. The central cross-section through the cryo-EM maps of CHIKV VLP (left) and Sindbis virus (Supplementary Fig. 2) are similar. The orientations of the icosahedral (2-, 3-, and 5-fold) axes as well as the quasi-threefold (q3) axis are shown with white lines. The map is calculated to 18 Å resolution.

Figure 3. Neutralization of CHIKV strains 37997 and LR2006 OPY-1 after DNA or VLP vaccination in mice and monkeys

(a) and (b) Sera from immunized BALB/c mice 10 days after the final immunization were tested with CHIKV strain 37997 (a) or LR2006 OPY-1 (b) E pseudotyped lentiviral vectors. Mice were immunized with the indicated 15 μg of DNA or 19 μg of VLP₃₇₉₉₇. Each C-E or E (strain 37997 and LR2006 OPY-1, respectively) plasmid was injected intramuscularly three times at 0, 3 and 6 weeks. VLP₃₇₉₉₇ with or without Ribi adjuvant was injected twice at 2 and 6 weeks. The experiment was performed in triplicate. The symbols show the average of the five mice and bars show the standard error of the mean. The curve fit was calculated by Prism software. (c) Rhesus monkeys were immunized with 20 μg of VLP₃₇₉₉₇ or PBS (control) intramuscularly three times at 0, 4, and 24 weeks. A neutralizing assay was performed with CHIKV strain 37997 (left panel) or LR2006 OPY-1 (right panel) E pseudotyped lentiviral vectors in sera collected from immunized monkeys at 10 days after each immunization. The symbols show the average of the six monkeys and bars show the standard error of the mean. (d) Neutralizing activity against CHIKV LR2006 OPY-1 in immunized monkeys’ sera after the 2nd and 3rd immunizations was confirmed by a standard plaque reduction neutralization test (PRNT). The symbols show the average of the six monkeys and bars show the standard error of the mean.

Figure 4. Protection against CHIKV LR2006 OPY-1 challenge in monkeys immunized with VLPs and in a CHIKV mouse model after passive transfer of purified IgG

(a) Monkeys injected with PBS (Control) or immunized with VLP₃₇₉₉₇ were challenged intravenously with 10¹⁰ PFU of the CHIKV strain LR2006 OPY-1 15 weeks after the final boost. The peak viremia at 24 h after challenge was measured by plaque assay. The detection limit was 1000 PFU per mL. Error bars represent the standard error of the mean. (b) The percentage of monocytes in the monkeys’ white blood cells was measured using a hematology analyzer before and 7 days after challenge with CHIKV. Error bars represent the standard error of the mean. An unpaired two-tailed t test was used for statistical analysis (Control at day 0 vs. 7, P = 0.0015; VLPs at day 0 vs. 7, P = 0.38; Control vs. VLPs at 7 days, P = 0.0036). (c) Purified IgG from a monkey immunized with VLPs (Immune) or a control monkey (Control IgG) was passively transferred into IFN-α/βR^-/- mice intravenously (2 mg of total IgG per mouse, n=5 per group). Recipient mice were challenged 24 h after IgG transfer with a lethal LR2006 OPY-1 challenge (30 PFU) by intradermal injection. The viremia in the mice after challenge was measured by quantitative RT-PCR (limit of detection = 40 RNA copies per mL). Error bars represent the standard error of the mean. (d) Survival curve of mice passively transferred with control IgG or CHIKV immunized IgG against lethal LR2006 OPY-1 challenge.