Qualitative Superiority, Not Magnitude, of Lung-Resident CD8+ T Cells Determines Protective Efficacy of Influenza NP Versus PB1 Vaccines Authors: Serin Lee, Eunju Jang, Jun Chang Published: 2026 ABSTRACT AbstractT cell-based universal influenza vaccines aim to generate robust lung-resident memory CD8+ T cells (TRM), yet it remains unclear whether conserved Ags elicit equally protective TRM pools. We compared recombinant adenoviral vectors expressing the influenza A nucleoprotein (rAd/NP) or polymerase basic protein 1 (PB1). After intranasal immunization, both induced comparable magnitudes of respiratory Ag-specific CD8+ T cells, yet only rAd/NP provided 100% survival after lethal influenza challenge. This disparity reflected qualitative differences in the TRM pool; nucleoprotein (NP)-specific cells were predominantly CD103−CD49a+, a phenotype associated with superior cytotoxicity, whereas PB1-specific cells were mainly CD103+CD49a−. Furthermore, NP-specific CD8+ T cells showed 100-fold higher functional avidity and stronger lung-local CTL activity than PB1-specific cells. Our findings demonstrate that TRM quality—specifically phenotypic bias and functional avidity—rather than magnitude, are the primary determinants of vaccine-mediated protection. This study underscores the critical importance of Ag selection in optimizing T cell-based universal vaccine strategies.Keywords: Universal influenza vaccine, Tissue resident memory T cells, Adenoviral vector, Nucleoprotein INTRODUCTION Influenza A viruses (IAV) remain a formidable threat to global public health, characterized by their ability to undergo rapid antigenic drift and shift (1). While current seasonal vaccines primarily rely on inducing neutralizing Abs against surface glycoproteins, their efficacy is often compromised by the emergence of mismatched or pandemic strains (2,3). Consequently, there is an urgent need for “universal” vaccine strategies that elicit broadly cross-reactive immunity. T cell-mediated immunity plays a critical role in targeting conserved viral Ags that are less susceptible to antigenic variation and eliminating infected cells at early stages of infection (4,5). Particularly, tissue-resident memory T cells (TRM) established at the site of infection, such as the lungs, can rapidly initiate localized immune responses upon Ag re-exposure, thereby enhancing protective immunity (6). Among the conserved internal proteins of IAV, nucleoprotein (NP) and polymerase basic protein 1 (PB1) are prominent targets for CD8+ T cell-mediated protection. Both Ags contain immunodominant epitopes capable of priming robust T-cell responses across diverse MHC backgrounds (7,8,9,10). However, it remains unclear whether these Ags are equally effective in generating a protective TRM pool. Recent studies have highlighted the importance of specific TRM subsets, particularly those defined by the expression of CD103 and CD49a. While CD103 facilitates epithelial retention, CD49a (the alpha subunit of the collagen-binding integrin α1β1) has been specifically linked to enhanced locomotion and superior cytotoxic effector programs within the lung parenchyma (11). Furthermore, functional avidity—the sensitivity of a T cell to its cognate peptide-MHC complex—plays a pivotal role in the early detection and elimination of virus-infected cells (12,13). Despite these insights, a comparative analysis of how different internal Ags influence the qualitative landscape of vaccine-induced TRM has not been fully explored. In this study, we utilized a recombinant adenoviral (rAd) vector platform to compare the protective efficacy and qualitative features of CD8+ T cell responses elicited by intranasal immunization with NP or PB1. We show that while both vaccines induce comparable numbers of Ag-specific CD8+ T cells in the respiratory compartments, only the recombinant adenoviral vectors expressing the influenza A nucleoprotein (rAd/NP) vaccine confers robust protection against lethal influenza challenge. By dissecting the TRM phenotype and functional avidity, we reveal that NP-specific T cells possess superior qualitative attributes—characterized by a CD49a-dominant phenotype and higher functional avidity—that underpin their superior protective capacity compared to PB1-specific cells. MATERIALS AND METHODS Animal experiments C57BL/6 female mice were purchased from Orient Bio Inc. (Seoul, Korea) and housed under specific pathogen-free conditions in the experimental facility at Ewha Womans University. All animal studies were conducted with the approval of the Ewha Womans University Institutional Animal Care and Use Committee (IACUC, No. 25-019). Construction of recombinant replication-defective adenovirus vaccine rAd/NP and rAd vector and the corresponding mock control had been generated previously, as described elsewhere (14). To construct the recombinant adenoviral vectors expressing the influenza A polymerase basic protein 1 (rAd/PB1), the influenza A PB1 gene was double digested with KpnI and XhoI and cloned into pShuttle-CMV vector. The resulting shuttle plasmid was linearized with PmeI and co-electroporated with pAdEasy-1 into electrocompetent BJ5183 cells to allow homologous recombination. The rAd plasmid was then linearized with PacI and transfected into HEK293 cells to amplify the virus. The recombinant adenoviruses were subsequently propagated in HEK293 cells and purified by double cesium chloride density-gradient ultracentrifugation, and viral titers were determined by 50% tissue culture infectious dose assay. Immunization and virus challenge C57BL/6 female mice were lightly anesthetized with isoflurane (Hana Biotech, Pyeongtaek, Korea) and intranasally immunized with rAd/NP (3×107 plaque-forming unit [PFU]) or rAd/PB1 (6×107 PFU-9×107 PFU) in 50 μl of PBS. Five weeks after vaccination, mice were challenged with A/Puerto Rico/8/34 (A/PR8, H1N1), which was provided by Dr. Baik Lin Seong (Yonsei University, Seoul, Korea). Cell preparation Mice were injected intravenously with 2 μg of PerCP/Cy5.5 anti-CD45 Ab (Clone 30-F11; BioLegend, San Diego, CA, USA) via the tail vein 5 min before euthanasia to distinguish tissue-resident cells from vasculature. Bronchoalveolar lavage (BAL) fluid was collected by flushing the airways with 1 ml of PBS through a catheter (BD Angiocath Plus, Tuas, Singapore). Lungs were perfused in situ with 5 ml of PBS containing 10 U/ml heparin, and digested with 20 U/ml of collagenase type IV (Worthington, Lakewood, NY, USA) and 100 μg/ml deoxyribonuclease I (Sigma-Aldrich, Irvine, United Kingdom), followed by incubation in RPMI 1640 (Welgene, Daegu, Korea) supplemented with 10% FBS at 37°C in 5% CO2 for 30 min. The digested tissue was then passed through a 70-μm cell strainer (SPL, Pocheon, Korea) to obtain single-cell lung suspensions. Spleens were dissociated through a 40-μm cell strainer in Iscove's Modified Dulbecco's Medium. Red blood cells in single-cell suspensions were removed by treatment with red blood cell lysis buffer (Sigma-Aldrich) for 5 min. Cell staining and flow cytometry Cells were incubated with rat anti-mouse CD16/CD32 (BD Pharmingen, San Diego, CA, USA) and 5 μg/ml streptavidin (Sigma-Aldrich) in FACS buffer (0.5% FBS, 0.1% NaN3 in PBS), for 20 min at 4°C to block Fc receptors. The cells were then stained for 30 min at 4°C with APC/Cy7 anti-mouse CD3 (clone 17A2; BioLegend), FITC anti-mouse CD8 (clone 53-6.7; BioLegend, clone KT15; MBL Life Science, Woburn, MA, USA), PE/Cy7-conjugated anti-mouse CD69 (clone H1.2F3; BioLegend), eFluor 450 anti-mouse CD103 (clone 2E7; Thermo Fisher Scientific, San Diego, CA, USA), Alexa Fluor 647 anti-mouse CD49a (clone Ha31/8; BD Biosciences, Milpitas, CA, USA), PE-labeled H-2Db/NP366-374 tetramer, and PE-labeled H-2Kb/PB1703-711 tetramer. After staining, cells were fixed with BD FACS™ Lysing Solution (BD Biosciences) at room temperature for 20 min. Intracellular cytokine staining Lung cells were stimulated with 1 μM NP366 or PB1703 peptide at 37°C in 5% CO2 for 5 h in the presence of BV421 anti-mouse CD107a (clone 1D4B; BD Biosciences) and brefeldin A (Thermo Fisher Scientific). For intracellular granzyme and perforin staining, lung cells were stimulated with 1 μM NP366 or PB1703 peptide for 24 h, and brefeldin A was added during the final 5 h of stimulation. After stimulation, cells were incubated with rat anti-mouse CD16/CD32 (BD Pharmingen) to block Fc receptors, followed by surface staining with APC/Cy7 anti-mouse CD3 (clone 17A2; BioLegend) and FITC or PerCP/Cy5.5-conjugated anti-mouse CD8 (clone 53-6.7; BioLegend, clone KT15; MBL Life Science), FITC anti-mouse CD103 (clone 2E7; Thermo Fisher Scientific), and Alexa Fluor 647 anti-mouse CD49a (clone Ha31/8; BD Biosciences). Cells were then permeabilized using the BD Cytofix/Cytoperm™ Fixation/Permeabilization Kit (BD Biosciences) and stained with PE anti-mouse IFN-γ (clone XMG1.2; BioLegend), PE anti-mouse Granzyme A (clone 3G8.5; BioLegend), PE/Cy7 anti-mouse Granzyme B (clone QA16A02; BioLegend), BV421 anti-mouse Perforin (clone S16009A; BioLegend) for 30 min at 4°C. CTL assay Splenocytes from naïve mice were used as target cells. For peptide-pulsed targets, cells were incubated with 10 μM NP366 or PB1703 peptide at 37°C for 2 h. After pulsing, peptide-loaded and unloaded target cells were differentially labeled with CFSE (Thermo Fisher Scientific) at 10 μM or 5 μM for peptide-pulsed cells and 0.5 μM for unpulsed cells, mixed at a 1:1 ratio, and subsequently stained with 5 μM PKH26 (MedChemExpress, Monmouth Junction, NJ, USA). For the in vitro CTL assay, lung cells from immunized mice were used as effector cells, and the effector-to-target ratio was adjusted to 5:1 by calculating the number of effector cells from the percentage of tetramer+ CD8+ T cells. Effector and target cells were co-incubated for 6 h at 37°C in 5% CO2. For the in vivo CTL assay, peptide-pulsed and unpulsed target cells were mixed at a 1:1 ratio, adjusted to a total of 2×107 cells in 200 μl, and injected intravenously. Mice were sacrificed 12 h after transfer, and the percentage of lysis was calculated as % lysis = (1 – pulsed/unpulsed) × 100. Functional avidity For soluble peptide stimulation, lung cells were incubated for 5 h at 37°C in 5% CO2 with NP366 or PB1703 peptides serially diluted to obtain final concentrations of 10-6–10-13 M. For EL4-presented peptide stimulation, EL4 cells were pulsed with serially diluted peptides at the indicated concentrations for 2 h to allow Ag presentation, and then co-cultured with lung cells at a 1:1 ratio for 5 h at 37°C in 5% CO2. Statistical analysis Statistical analyses were performed using GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA, USA). Data are presented as means ± SEM (n=4–9). Comparisons between two groups were assessed using unpaired two-tailed t-tests, and multiple group comparisons were analyzed using ordinary one-way ANOVA or two-way ANOVA, as appropriate. The p-values <0.05 were considered statistically significant. RESULTS AND DISCUSSION NP and PB1 vaccines elicit comparable magnitudes of respiratory CD8+ T cell responses but confers differential protection against lethal influenza challenge To achieve efficient mucosal immunity, we utilized an intranasal vaccination strategy. Mice were immunized with rAd/NP and rAd/PB1. Three weeks post-immunization, we differentiated lung parenchymal cells from the intravascular pool using intravenous anti-CD45 labeling prior to flow cytometric analysis (Supplementary Fig. 1). The frequencies of NP- and PB1-specific CD8+ T cells were comparable between the two groups. While absolute numbers of Ag-specific CD8+ T cells in the rAd/NP group trended higher, this did not reach statistical significance (Fig. 1A and B). The rAd/PB1 dose was adjusted based on the observation that a 1:1 formulation of rAd/PB1 and rAd/NP vaccines skewed the CD8+ T cell response toward NP at an approximate 1:2 ratio. To overcome this immunodominance and elicit CD8+ T cell responses of a comparable magnitude for both Ags, a 1:2 ratio of rAd/NP to rAd/PB1 was selected (Supplementary Fig. 2). Even after doubling the rAd/PB1 dose, the frequencies and absolute numbers of Ag-specific CD8+ T cells in both the BAL and lung parenchyma remained similar to the rAd/NP group (Fig. 1C-E). Collectively, these data indicate that rAd/PB1 is as capable as rAd/NP of eliciting a robust pool of Ag-specific CD8+ T cells within the respiratory compartments. Intranasal vaccination with rAd/NP or rAd/PB1 induces robust respiratory CD8+ T-cell responses, yet only rAd/NP protects against lethal PR8 challenge. Mice were immunized i.n. with rAd/NP or rAd/PB1, and Ag-specific CD8+ T-cell responses were assessed 3 weeks later. (A) Representative flow cytometry dot plots showing Db/NP366 tetramer+ and Kb/PB1703 tetramer+ cells in the lung after i.n. immunization with 3×107 PFU. (B) Frequencies % of CD8+ T cells and absolute numbers of Db/NP366 tetramer+ and Kb/PB1703 tetramer+ CD8+ T cells. (C) Representative dot plots showing Db/NP366 tetramer+ (NP group; 3×107 PFU) and Kb/PB1703 tetramer+ (PB1 group; 6×107 PFU) CD8+ T cells in BAL and lung. (D, E) Frequencies (D) and absolute numbers (E) of tetramer+ CD8+ T cells in BAL and lung. (F, G) At 5 weeks post-immunization, mice were challenged i.n. with influenza A/PR8 at 5 LD50. Body weight change (F) and survival rate (G) following PR8 challenge. Body weight is presented as percentage of starting weight and monitored daily for the indicated period; survival was recorded over the same period. Data are presented as mean ± SEM, based on n=4–9 mice per group. Statistical significance was determined using an unpaired two-tailed t-test. Survival curves were compared using the Gehan–Breslow–Wilcoxon test (p=0.0003). i.n., intranasally; LD50, lethal dose 50%; ns, not significant. We next evaluated whether the induced CD8+ T cells translate into functional protection. Vaccinated mice were challenged with a lethal dose of influenza A/PR8 (H1N1) five weeks post-immunization. Notably, the rAd/NP-immunized group showed only transient weight loss and achieved 100% survival. In stark contrast, the rAd/PB1 group exhibited no improvement in weight loss kinetics or survival compared to the PBS control, with all mice succumbing to the infection (Fig. 1F and G). This lack of protection in the rAd/PB1 group persisted even at higher vaccine doses, suggesting that the failure is not a simple dose-dependent phenomenon but rather a qualitative deficiency. Divergent TRM phenotypes and cytotoxic potential linked to protective efficacy Given the stark contrast in protection despite similar T cell magnitudes, we hypothesized that the vaccines drive distinct TRM compositions. While both vaccines induced high and comparable levels of the canonical TRM marker CD69, we observed significant differences in CD103 and CD49a expression (Fig. 2A). NP-specific CD8+ T cells were characterized by significantly higher CD49a expression, whereas PB1-specific cells were predominantly CD103+. Specifically, the NP group favored a CD103−CD49a+ TRM subset, while the PB1 group primarily formed a CD103+CD49a− population (Fig. 2B). Since CD49a+ TRM cells are known to be transcriptionally wired for superior cytotoxic effector programs, including granzyme B and perforin production, this phenotypic bias may underpin the superior efficacy of the rAd/NP vaccine (15). To evaluate the functional implications of these phenotypic differences, we assessed degranulation and cytotoxic effector profiles by measuring CD107a expression and cytotoxicity-associated molecules, including Granzyme A, Granzyme B, and perforin. The rAd/NP group showed a trend toward higher CD107a expression compared to the rAd/PB1 group, although this difference did not reach statistical significance, suggesting a more robust degranulation potential (Fig. 2C). To further determine whether this cytotoxic profile was preferentially associated with distinct TRM-related phenotypes, we analyzed cytotoxicity-associated marker expression within CD49a+ and CD103+ CD8+ T cell subsets after peptide stimulation (Fig. 2D). Notably, within the CD49a+ CD8+ T cell population, the rAd/NP group exhibited significantly higher frequencies of CD107a+, Granzyme A+, and Granzyme B+ cells than the rAd/PB1 group, whereas perforin expression was comparable between the two groups. In contrast, these cytotoxicity-associated molecules did not differ significantly between groups within either the CD103+ or CD49a+CD103+ double-positive CD8+ T cell subsets, further suggesting that the enhanced cytotoxic profile observed in the rAd/NP group was most evident in the CD49a+ CD8+ T cell population (Supplementary Fig. 3). Together, these findings indicate that CD49a expression marks a lung CD8+ T cell subset with enhanced cytotoxic programming, as reflected by increased degranulation and granzyme expression in the rAd/NP group, whereas CD103 expression was not associated with increased cytotoxic effector features. Phenotypic and functional characterization of TRM induced by rAd/NP and rAd/PB1 vaccination. Mice were immunized intranasally with rAd/NP (3×107 PFU) or rAd/PB1 (6×107 PFU), and analyses were performed at 3 weeks post-immunization. (A) Expression of TRM markers (CD69, CD103, and CD49a) on Ag-specific CD8+ T cells in BAL and lung. (B) Frequencies of CD103+CD49a− and CD103−CD49a+ subsets among Ag-specific CD8+ T cells in BAL and lung. (C) Frequencies of CD107a+ Ag-specific CD8+ T cells in the lung following stimulation with cognate peptide. (D) Frequencies of cytotoxicity-associated markers was analyzed in CD49a+ or CD103+ CD8+ T cell subsets after cognate peptide stimulation. Data are presented as mean ± SEM, based on n=4–5 mice per group. Statistical significance was determined using a two-way ANOVA with Šídák’s multiple comparisons test (B) and an unpaired two-tailed t-test (C, D). ns, not significant; GzmA, granzyme A; GzmB, granzyme B. *p<0.05; **p<0.01; ****p<0.0001. Based on the preceding findings that rAd/NP-induced CD8+ T cells may possess greater cytotoxic potential than those induced by rAd/PB1, we next examined whether the difference in protective efficacy between the groups was associated with actual killing capacity. We first performed an in vivo CTL assay to assess systemic cytotoxic activity (Fig. 3A). Because intravenously transferred target cells primarily home to the spleen, an intramuscular immunization model was used to establish robust splenic memory T cell responses and provide a reliable readout for systemic lysis. To complement this approach and assess lung-localized mucosal killing, we further performed an ex vivo CTL assay using effector cells isolated directly from the lungs (Fig. 3C). In vivo CTL activity measured in the spleen did not show a significant difference between the two vaccine groups (Fig. 3B). In contrast, local CTL activity in the lung was higher in the NP-specific CD8+ T cell group than in PB1 group, indicating enhanced cytotoxicity in the rAd/NP-immunized mice (Fig. 3D). Statistical analysis was omitted for the lung percent lysis data because effector cells were pooled within each group prior to the assay to accommodate limited cell yields. Ag-specific cytotoxic activity assessed by in vivo and in vitro CTL assays. (A) In vivo CTL assay scheme. Splenocytes from naïve mice were pulsed with the cognate peptide (NP366 or PB1703) or left unpulsed, labeled with CFSE (pulsed, 10 μM; unpulsed, 0.5 μM), and subsequently stained with PKH26 (5 μM). Labeled target cells (2×107 cells in 200 μl) were intravenously transferred into intramuscularly immunized mice at 3 weeks post-immunization, and spleens were harvested 12 h later for analysis. (B) Specific lysis (%) of peptide-pulsed target cells in the spleen in the in vivo CTL assay. Data are presented as mean ± SEM, based on n=4 mice per group. Statistical significance was determined using an ordinary one-way ANOVA with Šídák’s multiple comparisons test. (C) In vitro CTL assay scheme. Splenocytes from naïve mice were pulsed with the cognate peptide (NP366 or PB1703) or left unpulsed, labeled with CFSE (pulsed, 5 μM; unpulsed, 0.5 μM). At 4 weeks after intranasal immunization, lung cells were subjected to dead-cell removal and CD8+ T-cell enrichment, then co-cultured with labeled target cells for 6 h. (D) Specific lysis (%) of peptide-pulsed target cells in the in vitro CTL assay using lung-derived effector cells. Data shown are from pooled samples (n=4 mice per group). E:T, effector-to-target; ns, not significant. ****p<0.0001. Furthermore, we examined the functional avidity of the induced T cells. Lung-derived cells were stimulated for 5 h with serial dilutions of soluble peptide. NP-specific CD8+ T cells exhibited an approximately 100-fold lower effective concentration 50% (EC50) than PB1-specific cells across multiple assays, including direct peptide stimulation and EL4 target cell assays (Fig. 4A and B). When we repeated the assay using 2-fold serial dilutions instead of 10-fold dilutions, the separation between the two curves became more evident (Supplementary Fig. 4). This markedly higher functional avidity indicates that NP-specific CD8+ T cells can sense and respond to much lower concentrations of viral Ag, representing a superior qualitative feature that likely contributes to the observed protective immunity. Functional avidity of lung Ag-specific CD8+ T cells assessed by soluble peptide versus APC-presented peptide stimulation. Mice were immunized intranasally with rAd/NP (3×107 PFU) or rAd/PB1 (9×107 PFU), and lung cells were analyzed at 3 weeks post-immunization. Lung cells were stimulated for 5 h with NP366 or PB1703 peptides (10−6–10−13 M) either added directly or presented by EL4 cells pre-pulsed for 2 h; EL4 and lung cells were co-cultured at a 1:1 ratio. (A) Frequency of IFN-γ+ CD8+ T cells after stimulation with soluble peptide across the indicated concentrations. (B) Frequency of IFN-γ+ CD8+ T cells after stimulation with EL4-presented peptide across the indicated concentrations. Data shown are from pooled samples (n=3 mice per group). Conclusions The development of a universal influenza vaccine hinges on eliciting T cell-mediated immunity that is both robust and broadly cross-reactive. In this study, we demonstrated a striking disparity in protective efficacy between rAd/NP and rAd/PB1 vaccinations, despite both eliciting comparable magnitudes of Ag-specific CD8+ T cell responses in the respiratory tract. While rAd/NP-immunized mice achieved complete survival against a lethal H1N1 challenge, rAd/PB1-immunized mice remained entirely susceptible. This disconnect between the quantity and quality of the T-cell pool highlights that the mere presence of lung-resident CD8+ T cells is insufficient for protection; rather, the qualitative features—specifically the TRM phenotype and functional avidity—are the primary determinants of vaccine success. A key finding of our study is the divergent TRM subset composition driven by different internal Ags. Although both vaccines successfully induced CD69+ TRM cells, their expression of ancillary markers, CD49a and CD103, varied significantly. NP-specific CD8+ T cells were predominantly characterized by a CD103−CD49a+ phenotype, whereas PB1-specific cells were mainly CD103+CD49a−. Previous reports have suggested that CD49a expression on lung TRM cells is associated with enhanced cytotoxic potential and efficient locomotion within the infected tissue (15). In line with these reports, our subset-level analysis suggests that the CD49a-skewed response induced by rAd/NP is not merely a phenotypic difference, but is accompanied by a cytotoxicity-enriched effector profile characterized by enhanced degranulation- and granzyme-associated features. This finding provides a functional context for the CD103−CD49a+ phenotype observed in NP-specific cells. By contrast, the absence of comparable cytotoxic enrichment within the CD103+ or CD49a+CD103+ compartments suggests that CD103 expression was less closely associated with cytotoxic effector features in this setting. Together with the significantly superior lung-local CTL activity observed in the rAd/NP group, these findings support the interpretation that CD49a marks a TRM-like population with enhanced local effector readiness, although they do not establish CD49a as a causal driver of cytotoxic function. In contrast, the PB1-specific TRM pool, despite its presence in the lung parenchyma, may lack the requisite cytotoxic machinery or mobility to intercept viral replication effectively before the onset of lethal disease. Furthermore, the 100-fold difference in functional avidity between NP- and PB1-specific CD8+ T cells offers a possible explanation for the failure of the rAd/PB1 vaccine. High-avidity T cells are capable of detecting and responding to minute quantities of peptide-MHC complexes presented on infected cells early in the infection cycle (16,17). Importantly, functional avidity is a composite population-level readout influenced by multiple factors—including TCR–peptide MHC (pMHC) interactions, co-receptor expression, and downstream signaling sensitivity—rather than a direct measure of TCR–pMHC affinity alone. Among these determinants, CD8 co-receptor expression may be particularly influential under suboptimal TCR–pMHC interactions, as CD8 has been shown to enhance TCR–pMHC binding efficiency by modulating both association and dissociation kinetics (18). In our dataset, both vaccine groups displayed comparable CD8+ frequencies, yet the rAd/NP group exhibited significantly higher CD8 MFI (Supplementary Fig. 5), indicating higher per-cell CD8 expression. This observation is consistent with the possibility that higher CD8 expression may contribute to enhanced Ag sensitivity by stabilizing TCR–pMHC interactions and may facilitate proximal TCR signaling. However, repertoire-related determinants (e.g., precursor frequency, clonal diversity, and TCR affinity distributions) were not evaluated here, and their relative contributions should be addressed in future studies. Accordingly, the markedly lower EC50 values observed in NP-specific cells suggest that they can reach the threshold for activation and killing far more efficiently than PB1-specific cells when viral loads are still low. This rapid responsiveness is particularly critical in the context of influenza, where the window for preventing widespread alveolar damage is narrow. Given that TCR signal strength is a well-established rheostat for T cell differentiation and tissue-residency programs, differential TCR–pMHC signaling may represent a plausible contributor to this phenotypic divergence. This interpretation is further supported by the comparable magnitude of Ag-specific CD8+ T cell responses and the shared vector platform and inoculation route, which make gross differences in Ag availability or inflammatory milieu less likely. Interestingly, while systemic CTL activity in the spleen was comparable between the two groups, the divergence was only apparent at the local site of infection—the lungs. This dissociation is particularly relevant in respiratory viral infection, where viral replication initiates and rapidly amplifies within the airway and alveolar epithelium. This underscores the importance of evaluating “local” T cell quality rather than relying solely on systemic markers of immunity. The environment-specific enhancement of NP-specific CTL activity suggests that certain Ags may be better suited for mucosal vaccine strategies due to their ability to prime T cells with superior tissue-homing or resident-specific effector programs. In conclusion, our data suggest that NP is a superior target Ag for T cell-based influenza vaccines not because of its ability to induce more T cells, but because it elicits better T cells. NP-specific CD8+ T cells with higher Ag sensitivity and a predominance of cytotoxic CD49a+ TRM-like subsets may rapidly recognize and eliminate infected targets in the lung, even under low Ag-density conditions during early infection. By accelerating local viral clearance before substantial viral amplification occurs, these qualitative features could prevent the accumulation of viral burden that drives extensive alveolar damage. These findings suggest that future vaccine designs should prioritize the induction of high-avidity TRM responses, with CD49a expression serving as a marker associated with enhanced local effector capacity and protection against respiratory pathogens.