Madin-Darby canine kidney cells (MDCK II, ATCC^®, CRL-2936^TM) were used for the determination of viral titers with foci assay and plaque assay. The cells were purchased from the American Type Culture Collection (ATCC, Manassas, Virgina, USA) and cultured in Iscove’s Modified Dulbecco’s Medium (IMDM, Thermo Fisher Scientific, Waltham, Massachusetts, USA) supplemented with 10% (v/v) fetal bovine serum (FBS, Sigma-Aldrich, St. Louis, Missouri, USA), 1% (v/v) penicillin/streptomycin (Sigma-Aldrich). Cells were maintained in a 37°C and 5% CO₂ atmosphere. Human peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation from fresh blood of healthy human volunteers (registered in the biobank of the Department of Immunology at the Eberhard Karls University of Tübingen) and used to obtain human regulatory T cells. Ethical approval was granted after review by the Ethics Board of the Medical Faculty of the Eberhard Karls University of Tübingen and the University Hospital Tübingen (887/2020BO2).

Eight-week-old male C57BL/6 mice were obtained from the breeding facilities at the Friedrich-Löffler Institute (Federal Research Institute for Animal Health, Tuebingen, Germany) and were used for the survival study and analysis of immune cell populations. The mice had a body weight of 21 - 24 g at infection. The animals were fed with standard food which was available ad libitum. During this study, assessments included mortality checks, disease symptoms scores ( Supplementary Tables S1 , S2 ) and body weight. Mice were sacrificed as soon as their weight loss was above 25% compared to their initial body weight. Animal studies were approved by the Institutional Animal Care and Use Committee of Tuebingen.

Six-week-old female C57BL/6 mice (Charles River Laboratories, Sulzfeld, Germany) weighing 18 – 21 g at infection were used for the in vivo efficacy study of zapnometinib and standard of care treatment. The animals were fed with standard food which was available ad libitum Drinking water was available ad libitum. During this study, assessments included mortality checks, disease symptoms scores ( Supplementary Tables S3 , S4 ) and body weight. Mice were sacrificed by CO₂ gassing 6 days post infection (p.i.). The mouse study was reviewed and approved by the Regional Council Tuebingen (IM 01/21 G).

The pandemic influenza A virus (IAV) strain A/Regensburg/D6/09 (H1N1pdm09) was obtained from the Friedrich-Löffler Institute, Federal Research Institute for Animal Health, Riems, Germany, and was further propagated on MDCK II cells in a BSL2 laboratory. Shortly, a T-75 cell culture flask containing MDCK II cells was infected with a 1:100 IAV dilution in infection medium (IMDM medium supplemented with 0.2% BSA containing penicillin and streptomycin) for 1 h at 37°C. Afterwards, 10 ml of infection medium was added to the cells and incubated for 48 h. After incubation, the cell culture supernatant was centrifuged at maximum speed for 10 min to remove cell debris. The titer of the stock was determined, and aliquots were frozen at -80°C. For infection, mice were anesthetized by intraperitoneal (i.p.) injection of ketamine (10 mg/kg) solution or by 5% isoflurane in O₂. The mice were infected intranasally with either a low dose of 3 x 10³ plaque-forming units (pfu) or a high dose of 3 x 10⁵ pfu IAV in 50 μl PBS by inoculating 25 μl into each nostril. In the in vivo efficacy study, mice were infected with 3 x 10⁵ pfu IAV in 50 μl PBS by inoculating 25 μl into each nostril.

CI-1040 and zapnometinib were synthesized at ChemCon GmbH (Freiburg, Germany) and provided by Atriva Therapeutics (Tuebingen, Germany). Trametinib (GSK1120212) was purchased from ActiveBiochem (Hongkong, China). For the in vitro experiment, all inhibitors were prepared at a stock concentration of 10 mM in dimethyl sulfoxide (DMSO, Carl Roth GmbH, Karlsruhe, Germany) and further diluted in the respective media or buffer. For treatment of mice in the in vivo efficacy study, zapnometinib, oseltamivir (Hycultec GmbH, Beutelsbach, Germany) and baloxavir (Hycultec GmbH) were freshly dissolved in DMSO and then further diluted in 15% (v/v) Kolliphor EL (Sigma-Aldrich) and 80% (v/v) PBS. Each mouse received a dose of 25 mg/kg zapnometinib, 10 mg/kg oseltamivir, or 15 mg/kg baloxavir in an application volume of 100 μl of the formulation by oral gavage twice a day.

Virus titer was determined from lung homogenates of IAV-infected mice. Lungs were removed, weighed, and transferred into a Lysing Matrix D tube (MP Biomedicals Germany, Eschwege, Germany). Ice-cold PBS was added, the organs were shredded using the FastPrep^®-24 (MP Biomedicals) and centrifuged for 10 min at 13,000 × g at 4°C. The supernatant was stored at -80°C until used to perform an Avicel^® foci assay on MDCK II cells. Briefly, MDCK II cells were seeded in 96-well plates (Greiner Bio-One, Kremsmuenster, Austria) and incubated overnight. The cells were washed with PBS and infected with serial 3-fold dilutions of the lung tissue homogenates in triplicates. After 1 h p.i., the inocula were removed and the Avicel^® overlay medium [1.25% Avicel^® (FMC Biopolymer, Hamburg, Germany), 10% Minimum Essential Media (MEM, Thermo Fisher Scientific), 0.01% DEAE-Dextran (Sigma Aldrich), 2.8% NaHCO₃ (Merck, Darmstadt, Germany), 1% Penicillin/Streptomycin (Sigma Aldrich), 0.2% BSA (Carl Roth, Karlsruhe, Germany), 1% L- Glutamine (Sigma Aldrich)] was added for 24 h. The cells were fixed and permeabilized with PBS containing 4% Roti^® Histofix (Carl Roth). Afterwards, cells were immunostained with mouse anti-IAV nucleoprotein monoclonal antibody (Bio-Rad, California, USA) followed by goat anti-mouse IgG-HRP (Jackson ImmunoResearch Laboratories, Pennsylvania, USA). After washing, True Blue™ peroxidase substrate (SeraCare Life Science, Milford, USA) was added to detect foci, which were counted. The virus titer is given as the logarithm to the base 10 of the mean value. The detection limit for this test was<1.7 log₁₀ foci forming units (ffu)/ml.

To determine the reduction of viral titers in the in vivo efficacy study of zapnometinib and standard of care treatment, mice were sacrificed by CO₂ gassing on day 6 p.i. Lung homogenates in PBS were prepared as described above and serially diluted (5-fold dilutions starting at a 1:5 initial dilution) in IMDM medium supplemented with 0.2% BSA containing penicillin and streptomycin. The serial dilutions were added to MDCK II cells (seeded in 12-well plates (Greiner) 24 h before) and incubated for 1 h at 37°C. Subsequently, the inoculum was removed, the cells were overlaid with Avicel^® overlay medium and incubated for 48 h at 37°C with 5% CO₂. Afterwards, the Avicel^® overlay medium was removed, cells were washed twice with PBS, and fixed with 4% Roti^® Histofix in PBS for 30 min at 4°C before staining with 1% crystal violet. The plaques were counted, and the results were expressed as pfu/ml with a detection limit of 17 pfu/ml.

For flow cytometry analysis of immune cells of the lung or thymus, mice were infected with either the low or the high dose of IAV and sacrificed on day 3, 6, or 9 p.i. by CO₂ gassing. Organs were removed, minced, and lungs were digested in RPMI 1640 medium (Thermo Fisher) together with 0.125% collagenase II (Roche, Basel, Switzerland) for 30 min at 37°C and 5% CO₂. Then, the digested fractions were passed through a 70 µm BD cell strainer and washed three times. Cell pellets were resuspended in fluorescence-activated cell sorting (FACS) buffer (PBS containing 2 mM ethylenediaminetetraacetic acid (EDTA) and 2% (v/v) FBS) and stained with the required fluorochrome-conjugated antibodies at a dilution of 1:100 in FACS buffer for 30 min at 4°C. After incubation, cells were washed with FACS buffer, fixed, and analyzed by flow cytometry using FACS Calibur (BD Bioscience, Heidelberg, Germany). The following monoclonal anti-mouse fluorochrome conjugated antibodies were purchased from BD Bioscience and were used for the staining of immune cells: CD3 (CD3ϵ chain), CD4 (GK 1.5, H129.12), CD8a (Ly-2, 53–6.7.), CD25 (PC61), CD11b (M1/70), CD11c (HL3). The anti-mouse antibodies against CD244.2 (eBio244F4), NK1.1 (PK136), Ly6G (RB6–8C5) and foxp3 (FJK-16s) were purchased from eBioscience (NatuTec, Frankfurt, Germany). The following gating strategy was employed: First, lymphocytes were gated using an SSC-A/FSC-A lymphocyte gate, followed by FSC-A/FSC-H duplet exclusion (single cell gate). Subsequently the gate for CD3⁺ T cells was applied, followed by gating for CD4⁺ and CD8⁺ T cells. For the CD4⁺/CD25⁺FoxP3⁺ Tregs, the double positive CD25⁺FoxP3⁺ cell population was gated from CD4⁺T cells. The antibodies CD11b (M1/70), CD11c (HL3), CD244.2 (eBio244F4), NK1.1 (PK136), and Ly6G (RB6–8C5) were utilized for the detection of DCs, macrophages, neutrophils, NK and NKT cells. The detection of the mouse Tregs was performed with the mouse Regulatory T cell staining kit (Thermo Fisher Scientific) and the mouse Treg Detection kit (Miltenyi Biotec, Bergisch Gladbach, Germany) using CD4-VioBlue^®, CD25-APC, and anti-FoxP3-PE according to manufacturer’s protocol. The cells were analyzed by FACS Canto II (BD Bioscience) with the Diva software V.6.1.2 and for data analysis, FlowJo software (version 10, TreeStar Inc., USA) was used. The gating strategy of the mouse Tregs was carried out in several sequential steps. First, the lymphocytes were gated with a SSC-A/FSC-A lymphocyte gate, followed by the FSC-A/FSC-H duplet exclusion (single cell gate). Subsequently, dead cells were excluded with Aqua-Live Dead/FSC-A (live cell gate). The CD4⁺/CD25⁺ T cells were then gated from the living cells using CD25-APC/CD4- VioBlue^®. This double-positive population was then used to gate the CD4⁺/CD25⁺FoxP3⁺ cell population. For the detection of the Tregs isolated from human blood donors, the Treg Detection Kit with following fluorochrome conjugated antibodies was purchased from Miltenyi Biotec: CD45-VioBlue^®, CD4-VioGreen^®, CD25-VioBright^®, CD127-PE, anti-FoxP3-Vio667^®. The antibody staining was performed according to manufacturer´s protocol and the cells were analyzed by FACS Canto II (BD Bioscience) with the Diva software V.6.1.2. For data analysis, FlowJo software (version 10, TreeStar Inc., USA) was used. The gating strategy of the human Tregs was also carried out in several consecutive steps. First, the lymphocytes were delimited with an SSC-A/FSC-A lymphocyte gate, followed by FSC-A/FSC-H duplet exclusion (single cell gate) and dead cell exclusion (live cell gate). The enriched CD4⁺ T cells were then gated by FSC-A/CD4-VioGreen^®, followed by the CD45⁺CD25⁺ double-positive population, from which the gate for the FoxP3⁺ cell population finally followed.

For cytokine staining, a Th1/Th2 panel was utilized using the Bio-plex 200 system (Bio-Rad, Munich, Germany). Briefly, IAV infected mice were sacrificed on day 6 p.i. and serum samples of the mice were harvested. Therefore, blood was collected in serum sample tubes (Sarstedt, Nümbrecht, Germany) and centrifuged at 13,000 ×g for 5 min. These serum samples were added to the plate of the Bio-plex mouse cytokine Th1/Th2 kit and assayed for cytokine expression according to the manufacturer’s protocol. The measured cytokine concentration was then expressed in pg/ml by the instrument’s own software.

The in vitro cytotoxicity assay was used to detect cytotoxic cells (effector cells) in the mouse. In this method, radioactively labeled chromium-51 (⁵¹Cr) target cells are loaded with a specific IAV peptide. Specific lysis of the target cells caused by the effector cells and the radioactive amounts released as a result can be measured. For this purpose, lymphocytes were isolated from the lungs of IAV-infected mice by density gradient centrifugation and served as effector cells. These effector cells were serially titrated at 3-fold dilutions in MEM medium in a 96-well plate. MHC-I type H-2D expressing MC57 cells (Murine fibrosarcoma cell line, H-2b from a C57BL/6 mouse) served as target cells and were radiolabeled with 0.2 m Ci ⁵¹Cr for 1 h at 37°C and loaded with 10 μg/ml of the immunodominant IAV peptide (NP)_366-374 with the amino acid sequence ASNENVEIM. Subsequently, these target cells were co-incubated with the effector cells at effector to target cell ratios of 100:1, 30:1, or 10:1. Target cells in medium served as a control to measure spontaneous lysis (release of control) whereas target cells with HCl (Honeywell, Charlotte, North Carolina, USA) served as a control to measure the maximal target cell lysis and associated ⁵¹Cr release (maximum release). After 6 h incubation, supernatants were collected and measured for the presence of released ⁵¹Cr using the MicroBeta Counter 1750 (Wallac, Perkin Elmer, Waltham, Massachusetts, USA). The percentage of specific lysis was calculated according to the following formula [which was previously described in (34)]:

Ten donor mice were infected intranasally with a high dose of 3 x 10⁵ pfu IAV. In parallel, mice were previously chipped with an E-Mitter to monitor gross motor activity and temperature using the VitalView hardware and software system (Mini Mitter, USA). Gross motor activity and temperature were recorded every 5 minutes. A daily average is displayed for a period of 24 days. Mice were infected intranasally with a low infection dose of 3 x 10³ pfu IAV. Three of these mice served as acceptor mice. At day 4 p.i., donor mice were sacrificed, splenic regulatory T cells were isolated using the mouse CD4⁺/CD25⁺ Regulatory T Cell Isolation Kit (Miltenyi Biotec). 1 x 10⁶ isolated Tregs were subsequently applied intravenously (i.v.) to each of the three acceptor mice. In parallel, three chipped mice each received 0.6 mg/kg rapamycin (Merck) in PBS i.p. on day 4, 5, and 6 p.i. The remaining three mice served as controls without treatment. The health status of all chipped mice was monitored from the day of infection. Mice were weighed daily, and body temperature and disease symptoms score (DSS) were monitored until day 24 p.i.

To study the effect of MEK-inhibition on the induction of Tregs, regulatory T cells were isolated from human PBMCs using the CD4⁺CD25⁺CD127^dim/- Regulatory T cell Isolation Kit II and the Mini MACS Separator (Miltenyi Biotec) according to the manufacturer’s instructions. Briefly, the isolation of Tregs was performed in a two-step procedure. First, a cocktail of biotinylated antibodies and anti-biotin MicroBeads was used for depletion of non-CD4⁺ and CD127^high- cells. In the second step, the flow-through fraction of pre-enriched CD4⁺CD127^dim/- T cells were labeled with CD25 MicroBeads for subsequent positive selection of CD4⁺CD25⁺CD127^dim/- regulatory T cells. The isolated cells were resuspended in RPMI medium (RPMI 1640, Gibco) containing 10% serum, 1% penicillin/streptomycin, and 1% L-Glutamine with or without Interleukin-2 (R&D Systems, Minneapolis, Minnesota, USA), which served as a booster for the expansion of the cells to greater extend and were incubated for 20 min in the presence or absence of MEK inhibitors. Afterwards, cells were stimulated with anti-CD3/CD28-coated MACSiBeads (Miltenyi Biotec) for 36 h and measured by flow cytometry using FACS Canto II (BD Bioscience).

To examine the effect of zapnometinib and standard of care treatment on IAV infection, 20 female mice (5 per treatment group) were infected intranasally with 3 x 10⁵ pfu in 50 μl on day 0. Another 5 female mice served as uninfected control (Mock). At 24 h p.i., the mice were treated twice daily with either solvent control (5% (v/v) DMSO, 15% Kolliphor EL and 80% PBS) or 25 mg/kg zapnometinib, 10 mg/kg oseltamivir or baloxavir at a dosage of 15 mg/kg body weight, respectively. The compounds were administered via oral gavage with an application volume of 0.1 ml. The bidaily treatment was continued until day 3 p.i. and the study was terminated 6 days p.i. Lungs were harvested, dissociated into single-cell suspension using the Lung Dissociation Kit and the gentleMACS Dissociator (Miltenyi Biotec, program 37C_m_LDK_1) according to manufacturer´s protocol. The cell suspension was divided into two parts, one used to isolate Tregs using the mouse CD4⁺/CD25⁺ Regulatory T Cell Isolation Kit (Miltenyi Biotec) according to the manufacturer’s instructions and the other part was stored at -80°C until used for the determination of the virus titer by plaque assay.

Unless otherwise noted, data were collected in MS Excel, and statistical analyses were performed using GraphPad Prism software version 9 (GraphPad Software Inc.). Data from at least three independent experiments are presented as the mean ± standard deviation (SD). Sample size (n) is indicated in the figure legends. Statistical details for each experiment are described in the corresponding figure legends. Differences in viral titer and cell populations between low and high dose were analyzed by two-way ANOVA. Differences between the treatment groups were analyzed by one-way ANOVA. Significance in percent survival was analyzed using the log-rank test (Mantel-Cox). P values< 0.05 were considered statistically significant.

To investigate the influence of the amount of influenza A virus (IAV) used for infection on the severity of disease, C57BL/6 mice were intranasally infected with either a low dose (3 x 10³ pfu) or a high dose (3 x 10⁵ pfu) of IAV strain A/Regensburg/D6/09 (H1N1pdm09). Body weight, disease symptoms score (DSS), and survival were monitored for 20 days ( Figure 1A ). According to animal protection law, mice were sacrificed when body weight loss was > 25%. All mice started to lose weight 3 days p.i. Mice infected with the low dose of IAV reached a maximum body weight loss of 16% on day 8 p.i. Thereafter, the mice regained weight and reached to their initial body weight at day 20 p.i. ( Figure 1B blue squares). In contrast, mice infected with the high dose showed a more rapid and severe decrease in body weight from day 3 p.i. ( Figure 1B red circles). Finally, the manifestation of severe symptoms, including weight loss, culminated in mortality by day 12 p.i. at the high infection dose, while all mice infected with the low dose survived ( Figure 1C ). We then investigated whether differences in viral load in the lungs contributed to variations in disease progression. For this purpose, mice were infected with either the low or high dose of IAV and viral titers in the lungs were determined on day 3, 6, or 9 p.i. ( Figure 1D ). We found similar viral titers in the lungs of both doses on day 3 p.i. (low dose: 6.55 ± 0.35 ffu/ml, high dose: 6.47 ± 0.18 ffu/ml), on day 6 p.i. (low dose: 6.15 ± 0.18 ffu/ml, high dose: 6.73 ± 0.17 ffu/ml), and on day 9 p.i. (low dose: 4.09 ± 0.35 ffu/ml, high dose: 4.13 ± 0.10 ffu/ml) ( Figure 1E ). Thus, by using the two different doses for IAV infection, we were able to provide a sublethal and lethal influenza mouse model.

The fact that the virus titer was comparable after low and high dose IAV infection indicates that the viral load alone is not responsible for the development of severe clinical disease in mice. Therefore, the T cell response in mice after infection with either a low or a high dose of IAV was examined. The frequencies (in %) of CD4⁺ T cells, CD8⁺ T cells, and CD4⁺/CD25⁺FoxP3⁺ regulatory T cells (Tregs) in the lung and thymus were determined on day 3, 6 and 9 post infection ( Figure 2 ). Only on day 9 p.i. significantly higher levels of CD4⁺ T cells (low dose: 13.84 ± 2.34%; high dose: 25.65 ± 18.04%, Figure 2A ) and CD8⁺ T cells (low dose: 13.04 ± 3.21%; high dose: 26.07 ± 8.13%, Figure 2C ) were detected in the lung for both low dose (blue bars) and high dose (red bars) infection of IAV compared to uninfected controls (CD4⁺ T cells: 3.31 ± 0.57%; CD8⁺ T cells: 3.28 ± 0.59%, grey dots). At earlier time points (day 3 and 6 p.i.), the amount of CD4⁺ T cells (day 3: 8.58 ± 1.22%; day 6: 7.98 ± 3.15%, Figure 2A ) and the amount of CD8⁺ T cells (day 3: 5.63 ± 0.36%, day 6: 6.94 ± 0.55%, Figure 2C ) in the lungs of mice infected with the high dose was slightly increased compared to uninfected controls (dashed line), whereas the amount of CD4⁺ T cells (day 3: 2.42 ± 1.98%; day 6: 4.96 ± 3.19%, Figure 2A ) and the amount of CD8⁺ T cells (day 3: 2.16 ± 2.06%, day 6: 5.05 ± 3.72%, Figure 2C ) in the lungs of mice infected with the low dose was comparable to uninfected controls ( Figures 2A, C ). Significantly higher amounts of regulatory T cells in the lungs of mice infected with high dose of IAV were detected at day 6 p.i. (1.49 ± 0.46%) and day 9 p.i. (1.96 ± 0.62%, Figure 2E , red bars) compared to uninfected control. After a low infection dose, a significant increase in Tregs could be detected only on day 9 p.i. (2.02 ± 0.48%, Figure 2E , blue bar) compared to uninfected controls (0.51 ± 0.17%, Figure 2E , grey dots). In the thymus of low dose infected mice, no expansion of CD4⁺ T cells was observed (day 3: 9.45 ± 0.54%; day 6: 8.49 ± 2.52%; day 9: 9.42 ± 2.23%, Figure 2B , blue bars) compared to uninfected control (8.44 ± 1.08%, Figure 2B , grey dots). In contrast, a slight increase in CD4⁺ T cells after high infection dose was shown for all timepoints (day 3: 15.93 ± 0.94%; day 6: 17.04 ± 4.18%; day 9: 20.99 ± 12.04%, Figure 2B , red bars). No significant increase in the amount of CD8⁺ T cells was observed at any of the time points or infection doses (day 3: 2.62 ± 0.38% (low dose); 3.34 ± 1.07% (high dose); day 6: 2.54 ± 0.74% (low dose); 3.65 ± 1.81% (high dose); day 9: 4.29 ± 1.63% (low dose); 3.52 ± 1.47% (high dose), Figure 2D ) compared to uninfected controls. In contrast, the amount of Tregs in the thymus was altered after high dose (day 3: 0.57 ± 0.15%; day 6: 1.14 ± 0.29%; day 9: 3.32 ± 1.26%) but not after low infection dose (day 3: 0.23 ± 0.12%; day 6: 0.28 ± 0.07%; day 9: 0.27 ± 0.09%) compared to uninfected control (0.49 ± 0.15%). Significant differences were observed 6 and 9 days p.i. ( Figure 2F ).

To investigate the quality of the CD4⁺ T cell response in more detail, interferon-gamma (IFN-γ) and interleukin 4 (IL-4) cytokine response was analyzed on day 6 p.i. in the serum of mice infected with either a low dose or a high dose of IAV. Then, the IFN-γ/IL-4 ratio was determined to investigate whether the cells showed a Th1 or Th2 CD4⁺ T cell origin. A ratio of 1 and above indicates a Th1 response, while a ratio below 1 signals a Th2 response ( Figure 3 ). The serum IFN-γ concentration of infected animals was increased after both infection doses compared to control (4.27 pg/ml). A 2.5-fold higher concentration of IFN-γ was detected in mice infected with the low dose (87.75 pg/ml) than in animals infected with the high infection dose (32.88 pg/ml, Figure 3A ). In contrast, higher concentrations of IL-4 were detected in the serum of mice infected with the high dose (6.41 pg/ml) than in mice infected with the low dose (1.44 pg/ml, Figure 3B ). The ratio between IFN-γ and IL-4 at day 6 p.i. demonstrated that the low dose tended to result in a Th1 cytokine profile (ratio >1) while the high dose tended towards a Th2 cytokine expression ( Figure 3C ). In addition to IFN-γ and IL-4, the amount of IL-10 and tumor necrosis factor alpha (TNF) as representatives of a proinflammatory response were investigated. An increase of both cytokines was observed for both infection doses compared to the control (IL-10: 7.75 pg/ml; TNF: 22.89 pg/ml). IL-10 (low dose: 17.65 pg/ml; high dose: 21.35 pg/ml) and TNF (low dose: 253.44 pg/ml; high dose: 296.59 pg/ml), showed comparable concentrations in the serum of mice infected with the low and high infection dose ( Figures 3D, E ).

Next to the CD4⁺ T cell response, the cytotoxic T cell response was scrutinized in more detail. The proportion of IAV specific CD8⁺ T cells was quantified. Mice were infected with either low or high dose of IAV and sacrificed on day 6 p.i. A cell suspension was prepared from the lung cells and analyzed by FACS for the presence of CD8⁺ T cells that were specific for the immunodominant IAV peptide NP_366-374. Although the frequency of CD8⁺ T cells was similar between the low and high dose ( Figure 2C ), more pentamer-positive CD8⁺ T cells were identified after high dose infection (10.45 ± 0.55%) within, implying more virus specific CD8⁺ T cells after high dose infection ( Figure 3F ). The question now arose whether the higher amount of virus specific CD8⁺ T cells would also correlate with higher cytotoxicity of these CD8⁺ T cells. Therefore, an in vitro cytotoxicity assay was performed. Mice were infected with either a low or high dose of IAV and sacrificed on day 6 p.i. Lung lymphocytes were isolated and co-incubated with target cells labeled with the immunodominant IAV peptide NP_366-374 to determine the specific lysis of cytotoxic T cells. After infection with the low dose, a 10% higher cytotoxic T cell activity was observed compared to the high dose ( Figure 3G ).

Due to the different effector immune response in the lungs of mice infected with either low or high dose of IAV, one might speculate that CD4⁺/CD25⁺FoxP3⁺ Tregs inhibit the T cell effector response after infection with a high dose of IAV. Therefore, to better understand the impact of CD4⁺/CD25⁺FoxP3⁺ Tregs on influenza disease progression, Tregs were induced by high dose IAV infection. An adoptive transfer was performed by transferring these Tregs from donor mice into recipient mice, which were infected with the low dose of IAV. Furthermore, additional infected mice were administered rapamycin three times starting on day 4 p.i. Rapamycin itself is known to expand Tregs in vitro and in vivo (35). Mice infected with low dose of IAV served as controls. Figure 4A shows the scheme of the experimental procedure. Body weight, temperature and severity of disease were monitored daily. All infected mice lost weight from day 3 p.i. onward. Mice in the control group reduced their weight to a maximum of 85% by day 8 p.i., then regained weight and returned to initial weight by day 19 p.i. ( Figure 4B , gray line). Mice treated three times with rapamycin began to lose weight at the same time and reduced their weight to 80% by day 13 p.i. ( Figure 4B , orange line). These mice also regained weight and reached their initial weight at day 22 p.i. Mice that received a single transfer of Tregs 4 days p.i. also began to lose weight at day 3 p.i. and reached their minimal weight of 83% at day 12 p.i. ( Figure 4B , purple line). By day 21 p.i., they also regained their initial body weight. As mice have a loss in body temperature when mobility and activity drop down, body temperature was measured. The body temperature of the mice also started to decrease from day 3 p.i. onwards. In mice of the control group, body temperature decreased to a minimum of 34.7°C by day 8 p.i. In comparison, the body temperature of rapamycin-treated mice decreased to 33.7°C by day 11 p.i., and the body temperature of mice receiving Treg transfer decreased to 33.4°C by day 9 p.i. ( Figure 4C ). In addition, disease symptoms were monitored ( Figure 4D ). In all mice, the first disease symptoms appeared on day 4 p.i., a time when rapamycin treatment or adaptive transfer of Tregs began. Disease symptoms in mice from the control group increased by day 8 p.i., reaching a maximal score of 2. Subsequently, the mice recovered, and by day 13 p.i., these mice no longer showed disease symptoms ( Figure 4D , gray line). In mice treated with rapamycin, disease symptoms also reached a maximal score of 2 at day 8 p.i., but this score persisted until day 11 p.i. Thereafter, the mice began to recover slowly, but only on day 17 p.i. they showed no more symptoms ( Figure 4D , orange line). Transfer of Tregs also resulted in an increase in DSS and a maximum score of 2 at day 10 p.i. Disease symptoms persisted until day 17 p.i. ( Figure 4D , purple line). Compared to control animals, both treatment with rapamycin and transfer of Tregs resulted in a prolongation of disease duration. The results suggest that CD4⁺/CD25⁺FoxP3⁺ Tregs have an impact on pathogenesis by influenza viruses.

Based on these experiments one might argue that CD4⁺/CD25⁺FoxP3⁺ Tregs are involved in the pathogenicity and disease severity in IAV infected mice. We have previously reported that MEK-inhibition has an influence on the disease severity after IAV infection in mice (30). Thus, the influence of MEK-inhibition on CD4⁺/CD25⁺FoxP3⁺ Tregs was investigated in mice after IAV infection and in human PBMCs. CD4⁺/CD25⁺FoxP3⁺ regulatory T cells were isolated from human blood of healthy volunteers and stimulated in the presence or absence of the MEK-inhibitors zapnometinib, CI-1040 and trametinib for 36 h, followed by the quantification of CD4⁺/CD25⁺FoxP3⁺ Tregs via FACS analysis. Figure 5A shows that the formation of Tregs was significantly blocked in the presence of MEK-inhibitors. Treatment with CI-1040 reduced the amount of Tregs by 45.80 ± 19.75% compared to the untreated control. In cells treated with zapnometinib the amount of Tregs was decreased by 51.10 ± 24.55% and trametinib treatment reduced the amount of Tregs by 68.75 ± 14.06% ( Figure 5A ). These results demonstrate that MEK-inhibition reduces the formation of Tregs.

Next, the efficacy of zapnometinib to prevent the induction of Tregs was investigated in the IAV mouse model and compared with standard of care treatment for infection with IAV. Therefore, mice were infected with high dose (3 x 10⁵ pfu) of IAV and 24 hours later treated twice daily with 25 mg/kg zapnometinib, 10 mg/kg oseltamivir or 15 mg/kg baloxavir marboxil for 3 days. The mice were sacrificed on day 6 p.i., and the lungs were examined for the presence of Tregs. The study design is summarized in Figure 5B and a part of the gating in Figure 5C . IAV infection induced a significant higher amount of Tregs compared to non-infected Mock control (29.15 ± 6.39%, white bar, Figure 5D ). Treatment of mice with zapnometinib significantly reduced the induction of Tregs (by 53.02 ± 6.27%, grey bar) compared to control (black bar). In contrast, baloxavir (19.98 ± 36.09%, blue bar) had no effect on Treg induction, whereas oseltamivir resulted in an increase of Tregs (to 179.77 ± 58.15%, yellow bar). Thus, MEK-inhibitor treatment reduces CD4⁺/CD25⁺FoxP3⁺ Tregs from human PBMCs and in an IAV mouse model.