Vaccine order matters: mRNA priming with interferon signals boosts mucosal immunity in mice

When it comes to combining an intramuscular mRNA vaccine with an intranasal booster, the order in which the doses are given can make the difference between a robust, long-lived mucosal immune response and a comparatively weak one, according to a new mouse study published in EMBO Molecular Medicine. Researchers report that priming mice with an intramuscular mRNA vaccine against SARS-CoV-2 and then boosting them intranasally with an adjuvanted spike protein subunit vaccine generated markedly stronger and more durable mucosal and systemic immunity than the same two vaccines given in the reverse sequence. The key to this sequence dependence, the study finds, lies in the type I interferon response triggered by the mRNA vaccine at the time of priming.

The research team, led by Matthew J. Wiest and Pamela T. Wong of the University of Michigan together with Michael Schotsaert of the Icahn School of Medicine at Mount Sinai, set out to understand how heterologous prime-boost regimens—those combining different vaccine types and delivery routes—shape immunity depending on their chronological order. Parenterally administered vaccines against respiratory viruses are highly effective at preventing severe disease but poorly block infection and transmission, largely because they fail to induce strong mucosal immunity in the airways. Hybrid immunity, the combination of vaccination and natural infection, is known to generate particularly protective mucosal and systemic responses, and the researchers sought to emulate that profile through vaccination alone.

To do so, they used a rationally designed intranasal adjuvant consisting of a nanoemulsion combined with an RNA-based RIG-I agonist, referred to as NE/IVT, delivered with the spike protein of the original Wuhan Hu-1 SARS-CoV-2 strain. Mice received a prime-boost-boost schedule at three-week intervals, alternating between the BNT162b2 mRNA vaccine given intramuscularly and the adjuvanted or unadjuvanted spike subunit vaccine given intranasally. The team then profiled serum antibodies, airway IgA, neutralizing antibody titers against multiple variants, and T cell responses in the spleen, cervical lymph nodes, lungs, and nasal-associated lymphoid tissue.

The results showed that both the vaccine type and the route of the priming dose left a durable imprint on subsequent responses. Mice primed intramuscularly with mRNA and boosted intranasally with NE/IVT-adjuvanted spike produced dramatically higher levels of spike-specific IgA in bronchoalveolar lavage fluid than mice receiving the reverse sequence, exceeding even the additive effect of the corresponding homologous regimens. Antibody class switching also differed sharply: mice primed intramuscularly showed substantially higher IgG2c to IgG1 ratios, while intranasal priming skewed responses toward IgG1, suggesting that the priming modality durably shapes how booster doses are interpreted by the immune system.

Cellular immunity showed an even starker sequence dependence. The intramuscular mRNA prime followed by intranasal adjuvanted boost generated the highest frequencies of spike-specific CD8 tissue-resident memory T cells in the lungs and nasal-associated lymphoid tissue, along with robust CD4 effector memory cells in the cervical lymph nodes that drain the nasal mucosa. The reverse sequence, mucosal priming followed by systemic boosting, generated minimal or undetectable spike-specific CD4 effector memory cells in the cervical lymph nodes and few mucosal tissue-resident memory cells. Cytokine profiling revealed that the preferred sequence drove strongly Th1- and Th17-polarized responses—characterized by interferon-gamma, interleukin-2, tumor necrosis factor alpha, and interleukin-17A—whereas the reverse sequence produced a more Th2-skewed profile with elevated interleukin-5 and interleukin-13.

Protection was tested in a lethal challenge model using K18-hACE2 mice infected intranasally with the B.1.351 variant of SARS-CoV-2. Homologous intramuscular mRNA vaccination, homologous intranasal NE/IVT/spike vaccination, and heterologous regimens combining the two with the adjuvant all imparted sterilizing immunity in both the upper and lower respiratory tracts, with no detectable virus in the lungs or nasal turbinates at three or five days after infection. Heterologous regimens using unadjuvanted intranasal spike showed partial protection with faster viral clearance than unvaccinated controls, and the intramuscular prime followed by intranasal unadjuvanted boost performed better than the reverse order. Even among fully protected groups, lung cytokine profiles after challenge differed depending on the vaccination sequence, indicating that different cellular mediators contribute to protection in each case.

Durability emerged as another major advantage of the heterologous approach. When the researchers compared immune responses two weeks and three months after the final immunization, they found that serum neutralizing antibodies against the original virus and the BA.4/BA.5 variants were largely maintained across all regimens, with only the homologous mRNA group showing a decline against BA.4/BA.5. Mucosal IgA induced by the intramuscular mRNA prime followed by intranasal adjuvanted boost remained highly durable, and at three months was still higher than the levels achieved by either homologous regimen near the end of their immunization series. Cellular responses told a similar story: polyfunctional CD4 T cells and interferon-gamma-producing CD8 T cells waned markedly in the spleens and lungs of homologous mRNA-vaccinated mice over the three-month interval, but were maintained or even increased in the heterologous prime-pull group.

The mechanistic explanation centered on type I interferon. Acute-phase profiling showed that intramuscular mRNA immunization induced high serum levels of interferon-alpha, IP-10, and interleukin-6, and upregulated interferon-stimulated genes such as ISG15, Oas1a, and Mx1 both locally in the quadriceps muscle and distally in the nasal mucosa. In contrast, the intranasal NE/IVT/spike vaccine induced a robust interferon-stimulated gene response only locally in the nasal tissue, with minimal systemic interferon signaling. Intramuscular mRNA also increased the frequency and activation of dendritic cells in the draining lymph nodes, and prior mRNA priming enhanced the recruitment of antigen-presenting Ly6C-positive monocytes to the nasal mucosa when the intranasal booster was subsequently given.

To test whether the interferon response at priming was causally responsible, the researchers transiently blocked interferon-alpha receptor signaling with an antibody only during the mRNA prime. Blocking interferon signaling at priming reduced airway IgA, prevented the induction of spike-specific CD8 tissue-resident memory T cells in the lungs and nasal-associated lymphoid tissue, and shifted the polarization of responses in the mucosal compartments away from Th1 and Th17 toward Th2, while having minimal effects on splenic T cell responses. The blockade also altered antibody class switching, producing greater IgG1 skewing. These findings indicate that strong systemic type I interferon induction during priming conditions the mucosal environment to respond more effectively to later adjuvanted mucosal vaccination.

The study also explored whether established mucosal immunity could be reinforced after a prolonged delay. When mice were given an additional booster three months after their initial series, the heterologous prime-pull group showed further enhancement of airway IgA through either an intramuscular mRNA or an intranasal adjuvanted booster, suggesting that once strong mucosal B cell responses are established, they can be boosted through either route. In contrast, delaying the intranasal adjuvanted booster until three months after homologous mRNA vaccination failed to enhance airway IgA, demonstrating that mucosal boosting must occur relatively soon after systemic priming for optimal antibody induction, although delayed mucosal boosting did still establish mucosal Th1 and Th17 cellular responses in the cervical lymph nodes and lungs of previously mRNA-vaccinated mice. The authors note that the NE adjuvant has already demonstrated safety in multiple clinical trials, as have several RIG-I agonists, and that the findings in mice—particularly the outsized contribution of neutralizing antibodies in murine challenge models—will need further evaluation in additional models and ultimately in humans before informing human booster strategies.

Subject of Research: How the chronological order of intramuscular mRNA and intranasal adjuvanted subunit vaccination shapes mucosal and systemic immunity against SARS-CoV-2 in mice

Article Title: Spatiotemporal heterologous vaccination relies on type I interferons at priming to enhance mucosal immunity

Article References: Wiest, M. J., Laghlali, G., d’Acunzo, J., Chung, C., Karadag, D., Yan, V., Park, S.-C., Noureddine, M., Janczak, K. W., Chen, C., Farazuddin, M., O’Konek, J. J., García-Sastre, A., Baker, J. R., Jr, Schotsaert, M., & Wong, P. T. (2026). Spatiotemporal heterologous vaccination relies on type I interferons at priming to enhance mucosal immunity. EMBO Molecular Medicine. https://doi.org/10.1038/s44321-026-00539-w

Image Credits: AI Generated

DOI: 10.1038/s44321-026-00539-w

Keywords: heterologous vaccination, mucosal immunity, type I interferon, mRNA vaccine, intranasal booster, SARS-CoV-2, tissue-resident memory T cells, IgA, prime-boost schedule, NE/IVT adjuvant, mouse model, immune durability