<strong>система национального наблюдения за респираторными и энтерическими вирусами nrevss</strong>

Вирус коронавируса человека HKU1 (HcoV-HKU1) — это подтип коронавируса, который воздействует на людей. В этой статье будет рассмотрены образование и характеристики HcoV-HKU1, а также система мониторинга, используемая для отслеживания его циркуляционных паттернов.

Влияние предварительной кроссреактивной иммунности Т-клеток на COVID-19

Исследование иммунодоминантных T-клеточных эпитопов SARS-CoV-2

В данной работе мы впервые определили человеку релевантные иммунодоминантные T-клеточные эпитопы SARS-CoV-2 CD8+ и CD4+ в двух разных контекстах: иммунизация ДНК-вакцин, кодирующих белки SARS-CoV-2 шипучку (S), мембрану (M) или нуклеокапсид (N); и инфекция SARS-CoV-2. Затем мы установили кросс-реактивность OC43 вызванных экспрессии T-клеток к пептидам SARS-CoV-2, исследовали эффект предшествующего воздействия OC43 на последующую инфекцию SARS-CoV-2 и заболевания легких и определили вклад кросс-реактивных CD4+ T-клеток в кросс-защиту, вызванную OC43. Наши результаты показывают, что предшествующее воздействие OC43 вызывает кросс-защитный иммунитет, который частично посредством CD4+ T-клеток против инфекции SARS-CoV-2 и заболевания легких.

2023 Nov 17:S1341-321X(23)00284-2.

Online ahead of print.

Эпидемиология эндемичной инфекции человеческим коронавирусом во время пандемии COVID-19

J Infect Chemother

Аннотация

Методы: Носоглоточные мазки были взяты на наличие HCoV и SARS-CoV-2. Все медицинские данные были проведены ретроспективный анализ. Наиболее важной задачей было описать эпидемиологию HCoV в Фурано, Япония во время пандемии COVID-19. Второстепенной задачей было сравнить распространенность HCoV с SARS-CoV-2.

Результаты: С сентября 2020 по август 2022 года 113 (6,2 %) случаев из 1823 были положительными для любого HCoV. Активность HCoV-NL63 достигала своего пика в январе-марте 2021 года. Активность HCoV-OC43 достигала своего пика в июне-августе 2021 года. HCoV были обнаружены преимущественно у детей в возрасте ≤11 лет и наиболее часто в возрасте ≤2 лет. HCoV показали высокую детекцию в 2021 году, в то время как SARS-CoV-2 показали умеренную детекцию в 2020-2021 годах, но значительно увеличились в 2022 году.

Выводы: Во время пандемии COVID-19 активность HCoV-OC43 достигала своего пика летом. Частота заражения HCoV сильно варьировала в зависимости от возрастной группы и была выше среди лиц в возрасте ≤11 лет. Это отличается от предыдущих отчетов перед пандемией COVID-19. Эти результаты указывают на то, что динамика заболеваний HCoV остается неясной и продолжительное наблюдение является важным в пост-COVID-19 пандемии.

Ключевые слова: COVID-19; Эндемичный человеческий коронавирус; Эпидемиология; Пандемия; SARS-CoV-2.

Заявление о конфликте интересов

Заявление о конфликте интересов. Конфликтов интересов не выявлено.

Discussion

The level of protection mediated by pre-existing, cross-reactive T cells may be influenced by modulation of immunodominance hierarchies, as observed during Zika virus infection in humans and HLA transgenic mice with prior exposure to dengue virus74,136. Accumulating evidence suggests that immunodominance may be an important feature of the T cell response to SARS-CoV-2. For example, in a single cohort of SARS-CoV-2-infected individuals, researchers identified 2 populations of SARS-CoV-2 S protein-specific CD4+ T cells that were differentially activated: S751- and S236-specific, which were dominant and subdominant, respectively137. In addition, T cells recognizing an immunodominant seasonal HCoV/SARS-CoV-2 cross-reactive epitope restricted by HLA-B*1501 are observed in SARS-CoV-2 naïve and exposed individuals39. Similarly, CD8+ T cells recognizing the HLA-B*0702-restricted SARS-CoV-2 epitope N105-113 (the most immunodominant SARS-CoV-2 CD8+ T cell epitope identified to date) are present at high frequencies in unexposed healthy individuals25,28,79,80,81,82. Consistent with these human data, we found that CD8+ T cells specific for or cross-reactive with SARS-CoV-2 N104/105-113 epitope were immunodominant. Importantly, we observed that immunization of naïve HLA-B*0702 transgenic Ifnar1−/− mice with the single SARS-CoV-2 N104-113 peptide evoked a response that limited SARS-CoV-2-induced pathogenesis in the lung. Thus, the presence of immunodominant T cells within pre-existing SARS-CoV-2 cross-reactive T cell immunity could help to explain the disparate outcomes of COVID-19 patients.

In summary, the current study demonstrates a protective role for pre-existing seasonal HCoV/SARS-CoV-2-cross-reactive CD4+ T cell responses during SARS-CoV-2 infection. As human studies suggest both a protective and pathogenic role for HCoV-cross-reactive CD4+ T cell responses, a greater understanding of these T cell responses is required to facilitate T cell epitope-based rational vaccine design139. SARS-CoV-2 vaccines that are effective in individuals with antibody or B cell deficiencies are urgently needed to protect this highly vulnerable population and limit the emergence of new VOCs. Similarly, pan-HCoV vaccines that elicit cross-protective T cell responses represent key tools against new SARS-CoV-2 VOCs and HCoVs with pandemic potential.

Peer review

Nature Communications thanks Nimesh Gupta, Anthony Tan and the other, anonymous, reviewers for their contribution to the peer review of this work. A peer review file is available.

About this article

dos Santos Alves, R.P., Timis, J., Miller, R. et al. Human coronavirus OC43-elicited CD4+ T cells protect against SARS-CoV-2 in HLA transgenic mice. Nat Commun 15, 787 (2024). https://doi.org/10.1038/s41467-024-45043-2

Acknowledgements

We thank the Department of Laboratory Animal Care (Morag Mackay, Pascual Barajas, and Joseph Garza), the Department of Environmental Health and Safety (Dr. Laurence Cagnon and David Hall), and Flow Cytometry Core (Cheryl Kim) at the La Jolla Institute for Immunology for their assistance. We also thank Dr. Alessandro Sette at the La Jolla Institute for Immunology for providing HLA-B*0702 and DRB1*0101 Ifnar1−/− breeder mice. This study was funded by National Institutes of Health grants U19 AI142790-02S1 (to E.O.S. and S.S.) and U01 AI149644 (to R.S.B.), the GHR Foundation (to S.S. and E.O.S.), the Arvin Gottleib Foundation (to S.S. and E.O.S), and the Overton family (to S.S. and E.O.S.).

Data availability

All data supporting this study’s findings are included in the text, figures, supplementary material, and the source data provided in this paper. The complete dataset of immunofluorescence and histopathology images has been deposited in the repository: https://doi.org/10.5281/zenodo.10397796149. Source data are provided in this paper.

Rights and permissions

Reprints and permissions

Ethics declarations

The authors declare no competing interests.

Results

To investigate CD8+ and CD4+ T cell responses to SARS-CoV-2 in the context of human HLA alleles, we first identified the major predicted HLA-B*0702- and HLA-DRB1*0101-restricted T cell epitopes in SARS-CoV-2 S, N, and M proteins, which are known to be major CD8+ and CD4+ T cell targets in infected humans23. Using the Immune Epitope Database75 to identify potentially immunogenic peptides, we selected the top 1% of SARS-CoV-2 S, M, and N peptides predicted to have high-affinity binding to HLA-B*0702 or HLA-DRB1*010, and obtained 69 class I-restricted epitopes (Table 1) and 42 class II-restricted epitopes (Table 2). Mice from both strains were vaccinated with a DNA-based vaccine encoding SARS-CoV-2 S, M, or N proteins (Figs. 1A, B) on days 0 and 14, and spleens and lungs were collected 7 days later (Fig. 1C). Splenocytes and lung leukocytes were incubated with each peptide (vs no-peptide control), and IFNγ-producing peptide-specific T cells were quantified by ELISpot.

Table 1 Predicted HLA-B*0702-restricted epitopes from SARS-CoV-2 S-, M-, and N-proteins

Full size table

Table 2 Predicted HLA-DRB1*0101-restricted epitopes from SARS-CoV-2 S-, M-, and N-proteins

Fig. 1: Mapping SARS-CoV-2 S, N, and M protein-derived epitopes in DNA-vaccinated HLA-B*0702 and HLA-DRB1*0101 Ifnar1−/− mice.

Full size image

Splenocytes from DNA-vaccinated HLA-B*0702 transgenic mice produced significantly higher levels of IFNγ in response to 13 of the 69 peptides (S620-629, S678-688, S680-687, S680-688, S1056-1063, N64-74, N65-74, N66-74, N66-75, N66-76, N104-113, N105-113, and N105-114) compared to control cells; lung leukocytes from the same mice showed significantly elevated levels of IFNγ secretion in response to 7 of the 69 peptides (S1056-1063, N64-74, N65-74, N66-75, N66-76, N104-113, and N105-113; Fig. 1D). Of the 7 peptides that induced significant IFNγ responses in both lung leukocytes and splenocytes, the highest response was to the 3 peptides spanning residues 104 to 113 of the N protein (Fig. 1D); this was confirmed by intracellular cytokine staining (ICS) (Fig. S1). In vaccinated HLA-DRB1*0101 mice, 3 of the 42 predicted peptides (S315-329, S959-973, and M165-179) resulted in significant stimulation of IFNγ production by splenocytes, and 2 peptides (S315-329 and S998-1012) significantly stimulated a response in lung leukocytes; thus S315-329 stimulated splenocytes and lung leukocytes (Fig. 1E). In contrast to the class I-restricted response, none of the N protein-derived peptides stimulated a significant IFNγ response in DNA-vaccinated HLA-DRB1*0101 mice. These data demonstrate that SARS-CoV-2 DNA-based vaccines elicited T-cell responses dominated by recognition of S and N protein-derived peptides in the spleen and lung of HLA-B*0702 Ifnar1−/− mice and by S and M protein-derived peptides in HLA-DRB1*0101 Ifnar1−/− mice.

SARS-CoV-2 infection elicits effector CD8+ and Th1-biased CD4+ T cell responses in HLA transgenic Ifnar1 −/− mice

To determine whether the antigen-specificities of the T cell response elicited by SARS-CoV-2 DNA vaccines and live virus are similar, we infected HLA-DRB1*0101 Ifnar1−/− mice with mouse-adapted SARS-CoV-2 MA10 strain76 and HLA-B*0702 Ifnar1−/− mice with SARS-CoV-2 B.1.351 (Beta), which can replicate in mice without the need for adaptation77,78. Spleens were collected on day 8 post-infection (Fig. 2A), splenocytes were stimulated with select SARS-CoV-2 peptides, immunolabeled for cell surface markers, intracellular cytokines, and the degranulation marker CD107a, and the frequencies of activated (CD44+ CD62L−) effector CD8+ and CD4+ T cells were quantified by flow cytometry (Fig. S2).

Fig. 2: Mapping of SARS-CoV-2 S, N, and M protein-derived epitopes in SARS-CoV-2-infected HLA-B*0702 and HLA-DRB1*0101 Ifnar1−/− mice.

A Experimental protocol. HLA-B*0702 and HLA-DRB1*0101 Ifnar1−/− mice were infected with SARS-CoV-2 strains B.1.351 or MA10, respectively (104 PFU, IN), and spleens collected 8 days later. B, C ICS analysis of activated CD8+ T cells from B.1.351-infected HLA-B*0702 Ifnar1−/− mice (B) and activated CD4+ T cells from MA10-infected HLA-DRB1*0101 Ifnar1−/− mice (C). Splenocytes were stimulated for 6 h with the indicated 6 (B) or 42 (C) SARS-CoV-2 peptides (vs no peptide), immunolabeled for cell surface markers, intracellular cytokines, and the degranulation marker CD107a, and analyzed by flow cytometry. Data are presented as the mean ± SEM. N values: 6 (B) and 9 (C) mice/group, pooled from 2 independent experiments. Peptide vs control were compared using the nonparametric Kruskal–Wallis test. *P < 0.05; **P < 0.01; ***P < 0.001. Circles, individual mice. Blue bars on the x-axis, peptides that significantly stimulated CD4+ T cells with at least 1 secretion phenotype.

The CD8+ T cell response in B.1.351-infected HLA-B*0702 Ifnar1−/− mice was assessed by stimulating splenocytes with the 6 most potent SARS-CoV-2-derived peptides identified by DNA vaccination (S678-688, S1056-1063, N66-76, N104-113, M103-112, M164-172). While the frequencies of activated IFNγ+, IFNγ+/TNF+, IFNγ+/TNF+/IL-2+, and IFNγ+/CD107a+ CD8+ T cells were significantly increased in response to stimulation with epitope N104-113 (vs control), there was no significant expansion in response to the other 5 peptides (Fig. 2B, with the gating strategy represented in Fig. S2A). Thus, vaccination with a SARS-CoV-2 N protein-encoding DNA vaccine and direct infection with B.1.351 elicited effector CD8+ T cell response in HLA-B*0702 Ifnar1−/− mice that were both directed against the immunodominant epitope N104-113. This finding is in agreement with previous reports that SARS-CoV-2 N105-113 is the immunodominant epitope in SARS-CoV-2-infected individuals expressing HLA-B*070225,28,79,80,81,82.

As the class II-restricted response to SARS-CoV-2 DNA vaccines was lower in magnitude than the class I-restricted response, we stimulated splenocytes from MA10-infected HLA-DRB1*0101 Ifnar1−/− mice with each of the 42 SARS-CoV-2 peptides predicted to be immunogenic in the context of HLA-DRB1*0101, and then analyzed the frequencies of activated CD4+ T cells producing IFNγ alone or IFNγ and TNF (Th1 cells), IL-4 (Th2 cells), and IL-17A (Th17 cells) (Fig. 2C, with the gating strategy represented in Fig. S2B). All 42 peptides increased the frequency of IFNγ-producing cells compared with unstimulated control cells, but the increase was significant only in response to 2 peptides: S959-973 and N107-121. Three peptides expanded multifunctional IFNγ+/TNF+ CD4+ T cells (S315-329, S512-526, and N328-342). In contrast, none of the peptides stimulated CD4 + T cell production of IL-4 or IL-17A. Of note, N107-121 encompasses most of the immunodominant N104–113 CD8+ T cell epitope identified in both DNA-vaccinated and SARS-CoV-2-infected HLA-B*0702 Ifnar1−/− mice, and S315-329 also stimulated splenocytes from the DNA-vaccinated HLA-DRB1*0101 Ifnar1−/−mice. These findings demonstrate that the major CD4+ T cell response to primary infection with SARS-CoV-2 MA10 in HLA-DRB10101 Ifnar1−/− mice is Th1-biased, which is consistent with human studies showing that SARS-CoV-2 infection or vaccination elicits CD4+ T cells with a Th1-like phenotype23,83,84. Moreover, the agreement between the human studies and our findings in both DNA-vaccinated and SARS-CoV-2-infected HLA transgenic mice further validates these mouse models for investigation of human-relevant CD8+ and CD4+ T cell responses to SARS-CoV-2.

OC43 infection elicits CD8+ T cells with cross-reactivity to SARS-CoV-2 in HLA-B*0702 Ifnar1 −/− mice

The genomic sequence of SARS-CoV-2 N protein is 29% and 23% identical to the N protein sequences of β-coronaviruses (OC43 and HKU-1) and α-coronaviruses (NL63 and 229E), respectively85,86. To investigate whether exposure to seasonal common cold HCoVs can elicit CD8+ T cells that cross-react with SARS-CoV-2, we infected HLA-B*0702 Ifnar1−/− mice with OC43, the most common seasonal HCoV worldwide87,88. We then analyzed viral load in upper and lower airway tissues on days 1, 3, and 5 post-infection (Fig. S3A), and CD8+ T cell responses to SARS-CoV-2 peptides on days 1, 3, 5, 8, 16, and 30 post-infection (Fig. 3A). While OC43 genomic RNA levels in nasal turbinates increased between days 1 and 5, levels in lung were highest on day 1 and below the level of detection by day 5 (Figure S3B). Splenocytes prepared on days 8 and 16 post-infection with OC43 were stimulated with a panel of 37 HLA-B*0702-restricted SARS-CoV-2 CD8+ T cell epitopes previously shown to stimulate human CD8+ T cells based on IFNγ-ELISpot or ICS assays (NIAID Virus Pathogen Database and Analysis Resource; Table 3)25,89,90,91,92,93,94,95. These 37 peptides from SARS-CoV-2 M (n = 3), N (n = 7), ORF1ab (n = 24), ORF7 (n = 2), and ORF8 (n = 1) proteins—were selected to ensure that the analysis of OC43-elicited T cell reactivity was focused on the most human-relevant SARS-CoV-2 epitopes. The only significant activated CD8+ T response was at day 16 post-OC43 infection, and was focused on a single region in the N protein, with 9- and 12-fold expansion of N104-121-reactive IFNγ+ and IFNγ/TNF+ CD8+ T cells, respectively (Fig. 3B, with the gating strategy represented in Fig. S2C). To validate this finding, splenocytes and lung leukocytes isolated on day 8 post-infection were stimulated with the 69-peptide panel (Table 1 and Fig. 1C). Indeed, the frequencies of IFNγ-producing splenocytes and lung leukocytes were increased by only 3 SARS-CoV-2 peptides, all-encompassing the N104-113 epitope (Fig. 3C). Thus, exposure to OC43 elicits HLA-B*0702-restricted CD8+ T cells with reactivity to the SARS-CoV-2 N104-113 epitope. Of note, the SARS-CoV-2 and OC43 N104-113 sequences differ by a single amino acid residue (LSPRWYFYYL and LLPRWYFYYL, respectively), providing a basis for this cross-reactivity.

Fig. 3: Cross-reactivity of OC43-elicited CD8+ T cells for SARS-CoV-2 peptides.

Table 3 HLA-B*0702-restricted CD8+ T cell epitopes identified in the literature (May 2021)

OC43 infection elicits CD4+ T cells with cross-reactivity to SARS-CoV-2

To determine whether OC43 infection can also induce a CD4+ T cell response that cross-reacts with SARS-CoV-2, HLA-DRB1*0101 Ifnar1−/− mice were infected with OC43, and spleens and lungs harvested on days 8, 16, and 30 post-infection (Fig. 4A). Splenocytes and lung leukocytes isolated on day 8 were stimulated with the same 42-panel of SARS-CoV-2 peptides (Table 1) used to stimulate cells from DNA-vaccinated and SARS-CoV-2-infected mice (Figs. 1D and 2C, respectively). The frequencies of IFNγ-producing splenocytes were increased by 4 SARS-CoV-2 peptides (S54-68, S264-278, S758-772, and M32-46), and IFNγ-producing lung leukocytes were increased by 1 (S264-278) (Fig. 4B). To validate the CD4+ T cell cross-reactivity in this model, splenocytes were stimulated with a panel of 37 HLA-DRB1*0101-restricted peptides derived from SARS-CoV-2 E, S, M, N, ORF1ab, ORF3a, and ORF8 proteins (Table 4) previously shown to stimulate human CD4+ T cell responses by IFNγ-ELISpot, ICS, or MHC-binding assays18,90,96,97,98,99,100 (Fig. 4C). The gating strategy is represented in Fig. S2D. At day 8, IFNγ+ CD4+ T cells reactive with all 37 peptides were expanded in the spleen, although the increase was statistically significant only for cells stimulated with M66-80 and ORF3a116-130. At day 16, frequencies of IFNγ+ CD4+ T cells cross-reactive with SARS-CoV-2 ORF896-110 and ORF8101-115 were significantly increased. In contrast, polyfunctional SARS-CoV-2 cross-reactive IFNγ+/TNF+ CD4+ T cells were significantly expanded only in response to N86-100 and N261-275 peptides. In addition, while N86-100 and ORF3a108-120 peptides stimulated IL-2 production at day 16, none of the 37 peptides stimulated IL-4 expression (Fig. S3G). Finally, at day 30, the CD4+ T cell response was decreased for all peptides. Thus, exposure to OC43 elicits a Th1-biased CD4+ T cell response that cross-reacts with SARS-CoV-2 in the context of HLA-DRB1*0101. Viral load analysis on days 1, 3, and 5 post-infection (Fig. S3A) revealed levels of OC43 genomic RNA that were high in nasal turbinates on all 3 days and dramatically lower in lung (undetectable on day 1 and rising significantly but only slightly by day 3; Fig. S3C).

Fig. 4: Cross-reactivity of OC43-elicited CD4+ T cells for SARS-CoV-2 peptides.

Table 4 HLA-DRB1*0101-restricted CD4+ T cell epitopes identified in the literature (May 2021)

We next examined whether OC43 infection stimulates a SARS-CoV-2 cross-reactive antibody response in the context of HLA-DRB1*0101, given that CD4+ T cells play crucial roles in promoting and maintaining antibody responses101,102. To this end, serum was isolated from the OC43-infected mice at 6 time points from 0 to 100 days post-infection (Fig. S3D) and analyzed by ELISA for antibodies that bind to OC43 or SARS-CoV-2 S and N proteins. Anti-OC43 S IgG titers were detectable by day 14 and remained relatively stable up to day 100; in contrast, IgG reactive with SARS-CoV-2 S protein was not detected in sera of OC43-infected mice at any time point (Fig. S3E). IgG titers against the N proteins of both OC43 and SARS-CoV-2 were minimal at all time points (Fig. S3F). Thus, in our model of primary OC43 infection in HLA-DRB1*0101 Ifnar1-/- mice, OC43 S-specific IgG and SARS-CoV-2-reactive CD4+ T cells are present. These data demonstrate that prior exposure to OC43 elicits an HLA-DRB1*0101-restricted CD4+ T cell response that cross-reacts with SARS-CoV-2 epitopes.

Immunization with N104-113 peptide protects against SARS-CoV-2 infection and lung damage in HLA-B*0702 Ifnar1 −/− mice

To determine whether the OC43 cross-reactive CD8+ T cell response can protect against or exacerbate SARS-CoV-2 infection and/or pathology, HLA-B*0702 Ifnar1−/− mice were primed and boosted with N104-113 peptide on days 0 and 21, challenged with SARS-CoV-2 B.1.351 at 14 days post-boost, and tissues harvested 3 days later (Fig. 5A, with the gating strategy represented in Fig. S2E). In splenocytes from N104-113-immunized mice (vs mock-immunized), the frequencies of N104-113-reactive IFNγ+, polyfunctional (IFNγ+/TNF+ and IFNγ+/TNF+/IL-2+) and cytotoxic multifunctional (IFNγ+/CD107a+) CD8+ T cells were significantly increased (Fig. 5B), and lung appeared healthier by histopathology (Fig. 5C). In fact, quantification of histopathology data revealed 3 features of SARS-CoV-2-induced lung disease that were less pronounced (lower scores) in N104-113-immunized mice, although these differences were not significant: necrosis of bronchiolar epithelial cells (BEC), cellular debris in bronchioles, and suppurative bronchiolitis. When the Fig. 5A experiment was repeated with the MA10 strain of SARS-CoV-2 (Fig. S4A), our findings were confirmed: the N104-113-immunized mice had significantly higher frequencies of N104-113-reactive polyfunctional CD8+ T cells (Fig. S4B) and significantly lower histopathology scores (Fig. S4C). For the B.1.351-challenged mice, we also analyzed viral burden. Both RT-qPCR analysis of genomic RNA (Fig. 5D) and immunofluorescence analysis of SARS-CoV-2 N protein (Fig. 5E) revealed significantly lower levels of SARS-CoV-2 in lungs of N104-113-immunized mice. These results demonstrate that immunization of HLA-B*0702 Ifnar1−/− mice with SARS-CoV-2 N104-113 peptide elicits an antigen-specific polyfunctional CD8+ T cell response and protects against SARS-CoV-2 infection and lung disease.

Fig. 5: Protective effect of OC43 pre-exposure and SARS-CoV-2 N104-113 immunization on SARS-CoV-2 infection and lung disease in HLA-B*0702 Ifnar1−/− mice.

A Experimental protocol for B to E. Mice were injected with SARS-CoV-2 N104-113 vs DMSO (mock) on day 0 (complete Freund’s adjuvant, CFA) and again on day 21 (incomplete Freund’s adjuvant, IFA). Two weeks later, mice were challenged with SARS-CoV-2 B.1.351 (105 PFU, IN), and tissues were collected at 3 days post-challenge. Mice/group: 8 (peptide-immunized) and 7 (mock). B ICS analysis of activated CD8+ T cells. Splenocytes were stimulated for 6 h with SARS-CoV-2 N104-113 peptide, immunolabeled for cell surface markers, intracellular cytokines, and CD107a, and analyzed by flow cytometry. N = 7 and N = 8, respectively, for the mock and N104-113 groups. C Representative H&E-stained sections of lungs. Blue arrows, bronchiolar epithelial cells (BEC); black arrows, epithelial cells within bronchioles. Sections were scored from 0 (least severe) to 5 (most severe) for standard histopathological features of SARS-CoV-2-induced lung disease. D, E RT-qPCR of SARS-CoV-2 genomic RNA in lungs, and representative immunofluorescence of SARS-CoV-2 N protein (magenta) in lung sections with quantification of the N protein staining. In figure D, N = 7 and N = 8, respectively, for the mock and N104-113 groups. In figure E, N = 9 and N = 7, respectively, for the mock and N104-113 groups. F Experimental protocol for G to J. Mice were infected with OC43 (109 GE, IN) vs PBS (naïve) and challenged with SARS-CoV-2 B.1.351 (105 PFU, IN) 60–70 days later. Tissues were collected 3 days post-challenge. G ICS analysis of activated CD8+ T cells as described in B. Mice/group: 11 (OC43-infected) and 10 (naïve). H, I RT-qPCR of genomic SARS-CoV-2 RNA in lungs, and N protein staining in lung as described in D and E. Mice/group: 8 (OC43-infected) and 7-11 (naïve). J Lung histopathology and scoring as described in C. Mice/group: 8 (OC43-infected) and 11 (naïve). B, D, E, G–I Data pooled from 2 to 3 independent experiments and presented as the mean ± SEM. C, J Data pooled from 2 independent experiments and presented as violin plots. The means were compared using the two-sided Mann–Whitney test. Exact P values are indicated directly on the figure. Circles, individual mice.

Prior exposure to OC43 confers cross-protection against SARS-CoV-2 infection in HLA-B*0702 Ifnar1 −/− mice

Given that N104-113-immunization reduced SARS-CoV-2 burden and pathogenesis, we hypothesized that OC43-elicited CD8+ T cell immunity might be similarly protective. To test this, HLA-B*0702 Ifnar1−/− mice were infected with OC43 and challenged with SARS-CoV-2 on day 8 or 16 post-infection (Fig. S5A) or 60–70 days post-infection (Fig. 5F). Immunologic and virologic phenotypes were analyzed at 3 days post-challenge, which allowed us to focus on the effects of OC43-elicited immunity—rather than the primary T cell response to the SARS-CoV-2 challenge (primary antiviral T cell responses are generally not detectable until days 4 or 5 post-infection47,48,74).

To test for the presence of the OC43-elicited cross-reactive CD8+ T cell response in HLA-B*0702 Ifnar1−/− mice, splenocytes from OC43-exposed (vs naïve) SARS-CoV-2-challenged mice were stimulated with N104-113 peptide and analyzed by ICS (Fig. 5G). Polyfunctional CD8+ T cells were significantly increased in mice challenged at 60–70 days post-OC43 infection (Fig. 5G). In contrast, in mice challenged at 8 or 16 days, activated effector CD8+ T cell subsets expanded, but these increases were generally insignificant (Fig. S5B). RT-qPCR analysis revealed no effect of OC43 pre-exposure on SARS-CoV-2 genomic RNA levels in either lungs or nasal turbinates of mice challenged on days 8 or 16 (Fig. S5C). In contrast, lungs from mice challenged 60 to 70 days post-OC43 infection exhibited dramatic reductions in both SARS-CoV-2 genomic RNA (Fig. 5H) and N-protein immunoreactivity (Fig. 5I). While blinded histopathological analysis of lungs revealed no differences between OC43-infected and naïve mice challenged at 8 or 16 days post-infection (Fig. S5D), OC43-infected mice challenged at 60 to 70 days tended to have more bronchioles with clear lumina and viable epithelial cells lining the airway (i.e., proper polarization), and exhibited decreases in 3 histopathologic features (necrotic epithelial cells, cellular debris within bronchioles, and bronchiolar lesions), which were, however, not significant (Fig. 5J). Thus, a single prior intranasal exposure to OC43 can protect against SARS-CoV-2 infection in HLA-B*0702 Ifnar1−/− mice and may also limit SARS-CoV-2-induced lung damage in some mice. The immunologic and virologic data together suggest that OC43-elicited SARS-CoV-2-cross-reactive CD8+ T cells may contribute to the cross-protection against SARS-CoV-2 infection.

OC43 infection confers cross-protection against SARS-CoV-2 in HLA-DRB1*0101 Ifnar1 −/− mice in a manner partially dependent on CD4+ T cells

We have shown that HLA-DRB1*0101 Ifnar1−/− mice, mount an antigen-specific CD4+ T cell response against SARS-CoV-2 after DNA vaccination or viral infection (Figs. 1E and 2C), and a CD4+ T cell response to OC43 that cross-reacts with SARS-CoV-2 (Figs. 4 and S3G). To determine whether OC43 pre-exposure can protect against SARS-CoV-2, HLA-DRB1*0101 Ifnar1−/− mice were infected with OC43, challenged with SARS-CoV-2 16 days later (SARS-CoV-2 cross-reactive Th1 CD4+ T cell response peaked at 16 days post-OC43 infection; Fig. 4C), and lungs were harvested at 3 days post-challenge (Fig. 6A). Prior OC43 exposure led to dramatically lower levels of SARS-CoV-2 infection in the lungs, based on RT-qPCR analysis of SARS-CoV-2 genomic RNA and immunostaining for N protein (Fig. 6B, C), but did not significantly affect SARS-CoV-2 RNA levels in nasal turbinates (Fig. S6A). Histopathological analysis showed that the lungs of OC43-exposed/SARS-CoV-2-challenged mice were much healthier than those of naïve/SARS-CoV-2-challenged mice. Specifically, OC43 pre-exposure led to more bronchioles with clear lumina, viable epithelial cells, and significant improvement in all 5 features (Fig. 6D). Thus, OC43 pre-exposure for 16 days was sufficient to elicit a cross-protective response against SARS-CoV-2 infection and lung disease in HLA-DRB1*0101 Ifnar1−/− mice.

Fig. 6: Protective effect of OC43 pre-exposure on SARS-CoV-2 infection and lung disease in HLA-DRB1*0101 Ifnar1−/− mice.

A Experimental protocol for B to D. Mice were infected with OC43 (109 GE, IN) vs PBS (naïve), challenged with SARS-CoV-2 B.1.351 (105 PFU) 16 days later, and lungs collected at 3 days post-challenge. B, C RT-qPCR of SARS-CoV-2 genomic RNA, and representative SARS-CoV-2 N protein (magenta) immunofluorescence in sections with quantification of the N protein staining. N = 10 and N = 8, respectively, for the naïve and OC43 groups. D Representative H&E-stained sections of lungs. Blue arrow, bronchiolar epithelial cells (BEC); black arrow, epithelial cells within bronchioles; yellow arrow, perivascular cuffing. Sections were scored from 0 (least severe) to 5 (most severe) for standard histopathological features of SARS-CoV-2-induced lung disease. N = 10 and N = 8, respectively, for the naïve and OC43 groups. E Experimental protocol for F to H. Mice were infected with OC43 (109 GE, IN) vs PBS (naïve) and challenged with B.1.351 (105 PFU, IN) 16 days later. Mice were injected (IP) with CD4+ T cell-depleting antibody (α-CD4) vs isotype control antibody (ISO) once daily for 3 days before the challenge. Lungs were collected 3 days post-challenge. F, G RT-qPCR of genomic SARS-CoV-2 RNA in lungs, and N protein staining in the lung as described in B and C. N = 6 and N = 7, respectively, for the ISO and α-CD4 groups. H Lung histopathology and scoring as described for D. N = 6 and N = 7, respectively, for the ISO and α-CD4 groups. B–D, F–H Data, pooled from 2–3 independent experiments and presented as the mean ± SEM or violin plots, were compared using the two-sided Mann–Whitney test. Exact P values are indicated directly on the figure. Circles, individual mice.

The majority of SARS-CoV-2-specific T cell responses in humans are CD4+ T cells31,103,104. To determine whether the protective effects of OC43 pre-exposure were mediated by CD4+ T cells, we repeated these experiments in mice treated with a CD4 T cell-depleting antibody (vs isotype) immediately prior to SARS-CoV-2 challenge (Fig. 6E). Efficient CD4+ T cell depletion, confirmed by flow cytometry (Figure S6C), significantly reduced the protective effect of prior OC43 exposure on SARS-CoV-2 genomic RNA levels and N protein expression in lungs (Fig. 6F, G). However, CD4+ T cell-depleted, and isotype control mice (both OC43 infected) showed indistinguishable features of mild pneumonia in the lungs (Fig. 6H) and no difference in SARS-CoV-2 genomic RNA levels in nasal turbinates (Figure S6B). This indicates that OC43-elicited CD4+ T cells contribute to cross-protection against SARS-CoV-2 infection but do not significantly affect lung disease at day 3 after infection.

Author information

R.P.S.A., A.E.N., and S.S. conceived the study. R.P.S.A. and A.E.N. designed, performed, and analyzed the experiments. R.A., A.E.N., P.B.A.P., K.V., R.M., A.G., J.T., J.A.R.-N., M.N.N., E.M., S.L.-B., K.D., B.L., Z.M., S.M, N.S., and K.K. performed and analyzed experiments. S.R.L. and R.S.B. provided the MA10 virus strain. E.O.S. and S.S. obtained funding. S.S. supervised the research. R.P.S.A., A.E.N., and S.S. wrote the manuscript; other authors provided editorial comments.

Corresponding authors

Correspondence to Annie Elong Ngono or Sujan Shresta.