Tissue-resident memory T cells contribute to protection against heterologous SARS-CoV-2 challenge

Abstract

New vaccine formulations are based on circulating strains of virus, which have tended to evolve to more readily transmit human to human and to evade the neutralizing antibody response. An assumption of this approach is that ancestral strains of virus will not recur. Recurrence of these strains could be a problem for individuals not previously exposed to ancestral spike protein. Here, we addressed this by infecting mice with recent SARS-CoV-2 variants and then challenging them with a highly pathogenic mouse-adapted virus closely related to the ancestral Wuhan-1 strain (SARS2-N501YMA30). We found that challenged mice were protected from severe disease, despite having low or no neutralizing antibodies against SARS2-N501YMA30. T cell depletion from previously infected mice did not diminish infection against clinical disease, although it resulted in delayed virus clearance in the nasal turbinate and, in some cases, in the lungs. Levels of tissue-resident memory T cells were significantly elevated in the nasal turbinate of previously infected mice compared with that of naive mice. However, this phenotype was not seen in lung tissues. Together, these results indicate that the immune response to newly circulating variants afforded protection against reinfection with the ancestral virus that was in part T cell based.

Keywords: Adaptive immunity; COVID-19; T cells; Virology.

Figure 1. Neutralization against SARS-CoV-2 variants.

Mouse neutralizing antibody titers against SARS-CoV-2 variants were measured over time using an FRNT₅₀ assay. (A) Schematic of intranasal mouse infection and sera collection. (A–E) Variants included were (B) B.1.351, (C) BA.2.12.1, (D) BA.5, and (E) XBB.1.5. Naive mice were uninfected. (B) B.1.351 (n = 10) and naive (n = 5). (C) BA.2.12.1 (n = 7) and naive (n = 5). (D) BA.5 (n = 32) and naive (n = 5). (E) XBB.1.5 (n = 19) and naive (n = 5). Antibody titers were determined by the highest antibody dilution that resulted in a 50% reduction in the number of foci. 50% serum neutralization titers (NT₅₀) are listed above each group. Limit of detection (LOD) = 20 PFU. P values were measured by 1-way ANOVA followed by Tukey’s test for multiple comparisons. ***P < 0.001, ****P < 0.0001.

Figure 2. Sequential infection of mice with SARS-CoV-2 variants of concerns followed by SARS2-N501Y_MA30.

(A) Four-month-old (XBB.1.5) or 6-month-old (other SARS-CoV-2 variants) C57BL/6 mice were infected intranasally with SARS-CoV-2 variants or mock infected and were challenged with a lethal dose of SARS2-N501Y_MA30 3 months later. (B) Mice previously infected with B.1.351 (blue), BA.2.12.1 (orange), or XBB.1.5 (purple) or mock infected (PBS) were assessed for weight loss and survival after SARS2-N501Y_MA30 challenge. In some experiments, T cells were depleted at the time of challenge (depleted). B.1.351 data are from 1 experiment. PBS (n = 5) and B.1.351 (n = 4). BA.2.12.1 data are from 2 independent experiments. PBS (n = 9), PBS depleted (n = 4), BA.2.12.1 (n = 7), and BA.2.12.1 depleted (n = 5). XBB.1.5 data are from 2 independent experiments. PBS (n = 6), XBB.1.5 (n = 8), and XBB.1.5 depleted (n = 7). Red statistics denote PBS vs. XBB.1.5 depleted, black statistics denote PBS vs. XBB.1.5. P values were measured by log-rank followed by Bonferroni’s correction for multiple comparisons. (C) Sera obtained prior to challenge were tested for SARS2-N501Y_MA30 neutralizing antibodies. B.1.351 (blue), BA.2.12.1 (orange), BA.5 (green) or XBB.1.5 (purple), B.1.351 (n = 7), BA.2.12.1 (n = 7), BA.5 (n = 32), XBB.1.5 (n = 19), and naive (n = 5/group). Antibody titers were determined by the highest antibody dilution that results in a 50% reduction in the number of foci. Average titer is listed above each group. Data in B and C are shown as mean ± SEM. LOD = 20 PFU. P values measured by Mann Whitney U test (BA.2.12.1 and B.1.351) or 1-way ANOVA followed by Tukey’s test for multiple comparisons (BA.5). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 3. SARS-CoV-2–specific antibody binding in sera and nasal and lung tissues.

Mice were infected intranasally with 10⁵ PFU of XBB.1.5. (A) Schematic detailing the timeline and cohorts used for the antibody binding experiments. Three different cohorts of mice were used in this experiment. (B–D) Sera and nasal turbinate and lung tissues were harvested at the indicated times for the measurement of (B) total antibody, (C) IgG, (D) and IgA binding to WA1, BA.1, and XBB.1.5 full-length spike proteins, as described in Methods. (B) Antibody binding in serum. (C and D) Nasal turbinate and lung tissue IgG (C) and IgA (D) binding. For each cohort, n = 4. Data are from 1 experiment. All results were obtained prior to reinfection with SARS2-N501Y_MA30. P values were measured by 1-way ANOVA followed by Tukey’s test for multiple comparisons. Data in B–D are shown as mean ± SEM. LOD = 0.67. *P < 0.05, **P < 0.01.

Figure 4. Effect of T cell depletion on kinetics of virus clearance.

Four-month-old (XBB.1.5) or 6-month-old (other SARS-CoV-2 variants) C57BL/6 mice were infected with the indicated SARS-CoV-2 variant and challenged with SARS2-N501Y_MA30 3 months later. (A) Schematic detailing experimental timeline. CD4⁺/CD8⁺ T cells were depleted at the indicated time points. (B–D) Mice were initially infected with (B) BA.2.12.1, (B) B.1.351, (C) BA.5, or (D) XBB.1.5. PBS-treated mice were mock infected and then challenged with SARS2-N501Y_MA30. (B) BA.2.12.1- and B.1.351-infected groups were non–T cell–depleted mice (n = 4/group). Data represent 1 experiment. (C) 3 dpi: PBS (n = 8), BA.5 (non–T cell depleted), (n = 8 lungs, n = 10 nasal turbinates (NT), and BA.5 depleted (n = 9). 5 dpi: PBS (n = 9), BA.5 (n = 10), and BA.5 depleted (n = 10). Data are from 2 independent experiments. (D) The XBB.1.5-infected group comprised non–T cell–depleted mice. Mice were CD4⁺ T cell, CD8⁺ T cell or CD4/CD8⁺ T cell depleted. Each group contained 7–8 mice, from 2 independent experiments. Data in B–D are shown as mean ± SEM. Each symbol represents data obtained from 1 mouse. (E) XBB.1.5-infected mice were challenged with the B.1.1.7 (α variant). Virus titers in the lungs and nasal turbinates were measured at 5 dpi. Each group contained 4 mice. Data are from 1 experiment. (F) Lung pathology of XBB.1.5, XBB.1.5 T cell depleted, and PBS-treated mice at 5 dpi. PBS (n = 10), XBB.1.5 (n = 9), and XBB.1.5 infected and T cell depleted (n = 8). Evidence of edema is denoted by asterisks, and cellular infiltrates are marked with arrows. Scale bar: 450 μm (top) and 90 μm (bottom). All P values were measured by 1-way ANOVA followed by Tukey’s test for multiple comparisons. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Figure 5. Memory T cell characterization at 3 days after challenge.

Mice were challenged with SARS2-N501Y_MA30 3 months after XBB.1.5 infection. Mice were briefly treated with PerCP-Cy5.5–conjugated anti-Thy1.2 to label cells in the vasculature, as described in the Methods. Lungs and nasal turbinates were harvested for class I tetramer staining 3 days after reinfection and Thy1.2^– cells were analyzed by flow cytometry. PBS lungs (n = 8), PBS nasal turbinates (NT) (n = 7). XBB.1.5 lungs (n = 9), XBB.1.5 NT (n = 8). Data represent 2 independent experiments. P values were measured by Mann-Whitney U test. (A) Frequency (left) and number (right) of S539 tetramer⁺ T cells gated on the CD8⁺ T cell population. Representative plots for lungs (left) and nasal turbinates (right) are shown. (B) Frequency (left) and number (right) of virus-specific Trm cells gated on S539 tetramer⁺ T cells. Representative plots for lungs (left) and nasal turbinates (right) are shown. (C) Frequency (left) and number (right) of total Trm cells gated on the CD8⁺ T cell population. Representative plots for lungs (left) and nasal turbinates (right) are shown. Data for Trm frequency are shown as mean ± SEM. Data for Trm number are shown as geometric mean ± geometric SD. *P < 0.05, **P < 0.01, ***P < 0.001.