Addressing unexpected bacterial RNA safety concerns of E. coli produced influenza NP through CpG loaded mutant

Abstract

Influenza virus nucleoprotein (NP) is a promising target for universal influenza vaccines due to its conservation and high immunogenicity. Here, we uncovered a previously unknown factor that E. coli-produced NP carries bacterial RNA, which is crucial for its high immunogenicity but may pose safety and consistency concerns due to batch variability. To address these concerns, we developed a NP mutant (NP_mut) that lacks RNA binding activity but can be loaded with CpG1826, a synthetic oligodeoxynucleotide adjuvant that has been used in the FDA-approved Hepatitis B vaccine. The CpG1826-loaded NP_mut induced immune responses comparable to RNA-bound NP while eliminating safety risks. Additionally, the mixture of CpG1826-loaded NP_mut and 3M2e protein (three tandem copies of the ectodomain of influenza M2 protein) provided enhanced protection against influenza viruses challenge. Our findings highlight the adjuvant activity of bacterial RNA in E. coli-produced NP and propose a safer strategy for developing universal influenza vaccines.

Fig. 1. Co-administration of NP enhanced the 3M2e-specific immune responses.

A Scheme of the mouse experiment. Mice were immunized three times, and sera and bronchoalveolar lavage fluid (BALF) were collected to determine the antigen-specific antibodies. Splenocytes were collected for analysis of cellular immune responses. Lung tissue was collected for histopathological analysis. Mice (n = 5) were challenged with 5 × LD₅₀ of A/PR/8/34 (B) and 2.75 × LD₅₀ of A/SD/12/2019 (C) 2 weeks after the last immunization. Weight loss and survival rate were monitored daily for 14 days. The statistical significance of survival curves was determined by the log-rank (Mantel-Cox) test. ns: PBS versus NP, **PBS versus 3M2e/NP, *NP versus 3M2e/NP. M2e-specific (D) and NP-specific (E) immune responses were determined as described in the Materials and Methods (n = 3 to 5 per group). Data were shown as mean ± S.D. *p < 0.05, **p < 0.01, ***p < 0.001, and ****P < 0.0001, respectively (ANOVA).

Fig. 2. The adjuvant activity of NP on 3M2e relies on the NP-bound E. coli RNA.

A Non-denaturing PAGE showed that the NP protein, but not the 3M2e, contained bacterial nucleic acids. The nucleic acids and proteins were stained with EtBr and Coomassie blue, respectively. Recombinant NP was further confirmed by western blot using anti-NP serum. RNase-free DNase I (B) and DNase-free RNase A (C) treatments suggested that the recombinant NP proteins contained E. coli RNA. D–J The adjuvant activity of NP on 3M2e relies on the NP-bound E. coli RNA. Mice were immunized as shown in Fig.1A, and NP-specific (D) and M2e-specific (E) antibody titers in sera were shown. NP-specific (F) and M2e-specific (G) CD4⁺ T cells, and NP-specific CD8⁺ T cells (H) were determined as described in the Materials and Methods. I, J showed NP-specific and M2e-specific antibody titers in BALF, respectively. Data are presented as mean ± S.D, n = 4–5 mice for each group. “ns” indicates not significant. *p < 0.05, **p < 0.01, ***p < 0.001, and ****P < 0.0001, respectively (ANOVA).

Fig. 3. The NP-bound E. coli RNA enhanced the efficacy of the NP/3M2e vaccine.

Mice were immunized as shown in Fig. 1A, and challenged with influenza virus on day 42 post-vaccination. Weight loss and survival rate of mice (n = 5) were monitored daily for 14 days after challenged with 5 × LD50 of A/PR/8/34 (A). The statistical significance of survival curves was determined by the log-rank (Mantel-Cox) test. ns between PBS and NP-Tx, ** between PBS and NP, ** between PBS and 3M2e/NP, * between PBS and 3M2e/NP-Tx, ns between 3M2e/NP-Tx and 3M2e/NP, ns between NP-Tx and NP. Lungs (n = 3) were collected at 5 dpi for pathological analysis (B). Weight loss and survival rate of mice (n = 5) were monitored daily for 14 days after challenged with 2.75 × LD₅₀ of A/SD/12/2019 (C). The statistical significance of survival curves was determined by the log-rank (Mantel-Cox) test. ns between PBS and NP, ** between PBS and 3M2e/NP, * between NP and 3M2e/NP. Lungs (n = 3) were collected 4 dpi for pathological analysis (D). Representative results from each group were shown. Scale bars, 200 µm.

Fig. 4. An NP mutant with the CpG1826 but without the bacterial RNA binding activity as a target for universal influenza vaccines.

A Non-denaturing PAGE showed that the NP mutant (arginine mutates into glycine) lose the E. coli RNA binding activity. The RNase A-treated (NP-Tx) and untreated (NP) wild-type recombinant NP proteins were used as controls. The gel was stained with EB. B Non-denaturing PAGE showed that the recombinant NP_mut proteins could bind CpG1826 (highlighted in red box). NP, NP-Tx, and 3M2e were used as controls. C, D Co-delivery of CpG1826 enhanced the immune responses against NP_mut. Mice were immunized as shown in Fig. 1A. NP-specific total IgG, IgG1 and IgG2a in sera (C) (n = 5), and NP-specific IgG and IgA in BALF (D) (n = 3) were measured by ELISA. Data are presented as mean ± S.D. **, ***, and **** indicate p < 0.01, p < 0.001, and P < 0.0001, respectively (ANOVA). Two weeks after the last immunization, mice were challenged with 5 × LD₅₀ of A/PR/8/34. Body weight and survival rate of mice (n = 5) were monitored daily for 14 days (E). The statistical significance of survival curves was determined by the log-rank (Mantel-Cox) test. * between PBS and NP, ns between NP, NP-Tx/CpG1826 and NP_mut/CpG1826. Lungs (n = 3) were collected 4 days after challenge for histopathologic analysis (F). Scale bars, 200 µm.

Fig. 5. Co-delivery of CpG1826 enhanced the efficacy of NP_mut/3M2e.

Mice were immunized as shown in Fig.1A. NP-specific IgG (A) and M2e-specific IgG (B) were determined by ELISA (n = 5). NP-specific CD4⁺ T cells (C) and M2e-specific CD4⁺ T cells (D) were determined by activation-induced marker assay (n = 5). NP-specific CD8⁺ T cells (E) were determined by intracellular cytokine staining assay (n = 3). F NP- specific IgG and IgA in BALF (n = 5). G M2e-specific IgG and IgA in BALF (n = 5). Data are presented as mean ± S.D. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant (ANOVA). Mice (n = 5) were challenged with 5 × LD₅₀ of A/PR/8/34 (H) and 2.75 × LD₅₀ of A/SD/12/2019 (I) 2 weeks after the last immunization. Weight loss and survival rate were monitored daily for 14 days. Statistical significance of survival curves was determined by the log-rank (Mantel-Cox) test. H * between PBS and 3M2e/NP_mut, ** between PBS and 3M2e/NP_mut/CpG1826, * between 3M2e/NP_mut and 3M2e/NP_mut/CpG1826. I ns between PBS and 3M2e/NP_mut, * between PBS and 3M2e/NP_mut/CpG1826, ns between 3M2e/NP_mut and 3M2e/NP_mut/CpG1826.