A self-amplifying RNA vaccine prevents enterovirus D68 infection and disease in preclinical models

Abstract

The recent emergence and rapid response to severe acute respiratory syndrome coronavirus 2 was enabled by prototype pathogen and vaccine platform approaches, driven by the preemptive application of RNA vaccine technology to the related Middle East respiratory syndrome coronavirus. Recently, the National Institutes of Allergy and Infectious Diseases identified nine virus families of concern, eight enveloped virus families and one nonenveloped virus family, for which vaccine generation is a priority. Although RNA vaccines have been described for a variety of enveloped viruses, a roadmap for their use against nonenveloped viruses is lacking. Enterovirus D68 was recently designated a prototype pathogen within the family Picornaviridae of nonenveloped viruses because of its rapid evolution and respiratory route of transmission, coupled with a lack of diverse anti-enterovirus vaccine approaches in development. Here, we describe a proof-of-concept approach using a clinical stage RNA vaccine platform that induced robust enterovirus D68-neutralizing antibody responses in mice and nonhuman primates and prevented upper and lower respiratory tract infections and neurological disease in mice. In addition, we used our platform to rapidly characterize the antigenic diversity within the six genotypes of enterovirus D68, providing the necessary data to inform multivalent vaccine compositions that can elicit optimal breadth of neutralizing responses. These results demonstrate that RNA vaccines can be used as tools in our pandemic-preparedness toolbox for nonenveloped viruses.

Fig. 1.. Neutralizing antibody responses against EV-D68 are elicited by a repRNA vaccine coexpressing EV-D68 P1 and 3CD.

(A) Antigen-expression strategies from four repRNA candidates derived from the B1 US/MO/14–18947 isolate of EV-D68; P1 subunits are denoted by the colors red (VP4), blue (VP2), purple (VP3), and yellow (VP1). (B) Anti-VP1 Western blot analysis of BHK cell lysates harvested 24 hours after transfection with P1_IRES-3CD, P1_Δ3CD, VP1_HA2, and P1_T2A repRNAs. (C) Immunization schedule of female 6- to 8-week-old C57BL/6 mice (n = 5 per group) with repRNA vaccines. D, day. (D) Heat-inactivated serum samples, collected from immunized mice on day 42, were diluted and mixed with live homologous EV-D68 from clade B1, incubated, and transferred to monolayers of RD cells plated on xCELLigence e-plates. Impedance was measured every 15 min for 160 hours, and percentage neutralization was determined for each sample at each dilution. (E) Fifty percent neutralization was calculated for each sample, relative to virus-only and no-virus control conditions. (F) Forty-two days after prime, mice were challenged intranasally with 20 μl of a stock of mouse-adapted version (1 × 10⁸ PFU/ml) of the B1 subclade, US/MO/14–18949. Twenty-four hours later, mice were euthanized, and lungs were collected. Tissues were homogenized and centrifuged to remove cell debris, and supernatant was analyzed by plaque assay. Data in (D) are presented as individual values as well as sigmoidal 4PL-fitted curves. Data in (E) and (F) are presented as geometric means ± geometric SD. Dotted lines represent lower limit of detection.

Fig. 2.. Characterization of EV-D68 antigenic diversity using genotype variant–targeted repRNA-P1_IRES-3CD vaccines.

(A) To characterize EV-D68 antigenic relationships, six repRNA vaccines were constructed against a representative virus from each subclade of EV-D68, as outlined in Table 1 and fig. S2. Female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized intramuscularly on days 0 and 28 with 1 μg of a corresponding repRNA/LION vaccine (A1, A2, B1, B2, B3, or C repRNAs), and serum samples were collected on day 42 for measurement of vaccine-homologous and vaccine-heterologous neutralization by RTCA in vitro. Horizontal dotted lines represent the upper (>5120) and lower (<40) limits of detection. (B) To evaluate short-term homologous protection in vivo, female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized with 1 μg of B1 repRNA/LION and challenged intranasally with 3 × 10⁵ PFU of B1_MA on day 41. (C and D) The next day, mice were euthanized, and nasal turbinates (C) and lungs (D) were harvested for evaluation of viral load by plaque assay. (E) To evaluate long-term, dose-dependent efficacy after homologous or heterologous challenge, female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized with 1 or 0.1 μg of B3 repRNA/LION and challenged intranasally with either B3 or B1_MA on day 183. (F and G) The next day, mice were euthanized, and nasal turbinates (F) and lungs (G) were harvested for evaluation of viral load by plaque assay. Data in (A), (C), (D), (F), and (G) are presented as geometric means ± geometric SD, and dotted lines represent lower limit of detection. *P < 0.05 in (C) and (D) as determined by log₁₀-transformed Student’s t test and in (G) by log₁₀-transformed two-way ANOVA with Bonferroni’s multiple comparisons test.

Fig. 3.. Heterologous immunization incompletely protects against EV-D68 infection but alleviates host inflammatory responses to infection.

(A) Female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized intramuscularly on days 0 and 55 with 2 μg of corresponding repRNA/LION vaccine (A1, B1, or C repRNAs or a repRNA encoding RSV fusion antigen as a negative control). Serum samples were collected on days 27 and 95, and on day 102, animals were challenged intranasally with 3 × 10⁵ PFU mouse-adapted B1 (B1_MA). The next day, lungs and nasal cavities (NCs) were harvested. (B) Sera from days 27 and 95 were measured for B1_MA neutralizing antibody responses by RTCA in vitro. Horizontal dotted lines represent the upper (>1280) and lower (<40) limits of detection. (C) Viral loads in lungs and NCs were measured by plaque assay. (D to G) Host responses to B1_MA infection of the lung were measured using NanoString (nCounter mouse host response panel) to quantify host mRNA transcripts in RNA extracted from homogenized lung tissue collected from the B1-, C-, and RSV-vaccinated groups and a naïve uninfected group (n = 3 per group). (D) Differential gene transcripts between B1- and C-vaccinated mice or between B1-, C-, or RSV-vaccinated mice and uninfected mice are shown. Horizontal lines represent a P value of <0.005, and vertical lines represent fold change of >3 or −3. Numbers in top right quadrants indicate the number of gene transcripts in the >3-fold and P < 0.005 cutoffs. (E and F) NanoString-computed (E) macrophage or (F) neutrophil cell abundance scores are shown. (G) Differential gene transcript expression between the RSV- and combined B1/C-vaccinated groups is shown, highlighting the top nine host response mRNAs preferentially up-regulated in the RSV-vaccinated group compared with those in the B1/C-vaccinated groups. Data in (B) and (C) are presented as geometric means ± geometric SD. Data in (E) and (F) are presented as means ± SD. Data in (D) and (G) are presented as differential expression volcano plots. *P < 0.05 in (B) and (C) as determined by log₁₀-transformed Kruskal-Wallis test; in (B), *P < 0.05 for the B1 repRNA group at each time point compared with all other groups at each time point. P values in (D) and (G) as determined by Rosalind.bio. *P < 0.05 in (E) and (F) as determined by log₁₀-transformed one-way ANOVA.

Fig. 4.. RepRNA formulation drives differential protective outcomes in upper and lower airways of mice after heterologous EV-D68 challenge.

(A) Female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized with 1 or 10 μg of B2-repRNA formulated in LION or LNP, 35 days apart, and challenged intranasally with 4.2 × 10⁶ PFU of B1_MA on day 49. (B) Serum samples were collected on days 28 and 49 and assayed for B2 neutralization activity by RTCA in vitro. (C and D) On day 50 after prime (1 day after challenge), mice were euthanized, and nasal turbinates (C) and lungs (D) were harvested for evaluation of viral load by plaque assay. Data in (B) are presented as violin plots with minimum, maximum, quartiles, and median; data in (C) and (D) are presented as geometric means ± geometric SD with dotted horizontal lines representing the lower limit of detection.*P < 0.05 in (B) to (D) as determined by log₁₀-transformed two-way ANOVA with Bonferroni’s multiple comparisons test.

Fig. 5.. Passive transfer of serum from B2-repRNA/LION–vaccinated mice protects 8-day-old IFNAR^−/− mice against paralytic disease caused by EV-D68 infection.

(A) Female C57BL/6 mice, 6 to 8 weeks old (n = 5 per group), were immunized with 3 μg of LION/repRNA encoding either B2 antigen or HA from influenza virus (flu) as a negative control 28 days apart and terminally bled on day 42. Seven-day-old type I interferon receptor knockout (Ifnar1^−/−) mice then received pooled B2-vaccinated (n = 5) or flu-vaccinated serum (n = 7) or PBS (n = 4) by intraperitoneal injection (75 μl). One day later, animals were challenged with B2 virus and monitored for clinical signs of disease for 21 days. (B) Day 42 serum was assayed for B2 neutralization activity by RTCA in vitro. (C to E) After challenge, animals were monitored for survival (C), weight gain (D), and total neurological score (E). (F) Bilateral hindlimb flaccid paralysis (red arrows) was present in the sole surviving mouse that received flu-repRNA/LION serum as compared with a representative animal that received B2-repRNA/LION serum.

Fig. 6.. B1 repRNA/LION is immunogenic and elicits broadly neutralizing antibody responses in rhesus macaques.

(A) Immunization and bleed schedule. Male (n = 4) and female (n = 8) rhesus macaques, 2 to 12 years of age, were randomized and distributed into two groups (n = 6 per group) based on gender and age (table S1). All animals received 25 μg of B1 repRNA/LION or a control repRNA encoding Crimean Congo hemorrhagic fever virus antigen (irrelevant control) on days 0 and 42. (B) Serum samples were collected on days 0, 14, 28, 42, 49, and 63, and B1 neutralizing antibody responses were assayed by real-time cell analysis. (C) To measure breadth of the neutralizing antibody response, the day 63 serum samples were assayed against A1, A2, B1, B2, B3, and C viruses by real-time cell analysis. Data in (B) are presented for each individual animal; data in (C) are presented as violin plots with minimum, maximum, quartiles, and median.