Dimethyl Sulfoxide Modulates T Cell–Mediated Immunity to Attenuate Influenza A Virus Infection Authors: Po-Hsuan Jeng, Meng-Wei Tsai, Shing-Hwa Huang, Tien-Ru Huang, Mei-Fei Chen, Gu-Jiun Lin, Shao-Cheng Liu Published: 2026 ABSTRACT PurposeInfluenza A virus (IAV) remains a major threat because vaccines provide limited cross-protection and antiviral resistance is rising. Dimethyl sulfoxide (DMSO) is immunomodulatory, but its effect on IAV has not been tested in vivo. We evaluated its efficacy and mechanism in a murine H1N1 model.MethodsMale C57BL/6 mice were intranasally challenged with an LD50 of A/Puerto Rico/8/1934 (H1N1) and given daily subcutaneous DMSO (0.22 g/kg/day) or phosphate-buffered saline (PBS) on days 1–7. Survival was monitored for 21 days; lung and bronchoalveolar lavage fluid (BALF) were collected on day 8. Lung histology (hematoxylin and eosin), BALF cytokines (sandwich ELISA), lung and splenic IFN-γ–producing CD4+ and CD8+ T cells (flow cytometry), and whole-lung RNA sequencing (differential expression, KEGG enrichment) were assessed. BALF viral burden was measured by neuraminidase activity assay and TCID50. In vitro, splenic CD8+ T cells exposed to 0%, 0.5%, or 1.0% DMSO were immunoblotted for lymphocyte-activation gene 3 (LAG3).ResultsDMSO significantly improved 21-day survival and attenuated alveolar inflammatory infiltration. Day-8 BALF viral titers were significantly lower with DMSO (mean 2.2×103 vs 4.3×104 TCID50/mL; p = 0.034), as was neuraminidase activity. BALF IFN-γ and TGF-β were elevated, whereas TNF-α, IL-6, and IL-1β were not. IFN-γ+ CD8+ T cells increased in the lung but not the spleen; CD4+ populations were unchanged. RNA sequencing showed enrichment of cytokine–cytokine receptor interaction and PI3K/Akt signaling and upregulation of T-box factors, PPAR-γ, and Nek kinases. In vitro, DMSO accelerated LAG3 induction in CD8+ T cells within 12–48 h.ConclusionDMSO is associated with attenuated IAV severity, a lower day-8 airway viral burden, and a localized expansion of IFN-γ–producing CD8+ T cells, without a concomitant rise in the pro-inflammatory cytokines TNF-α, IL-6, or IL-1β. These associations are correlative and support further investigation of DMSO as a host-directed adjunct for influenza. Introduction Influenza is a highly contagious respiratory disease that exerts a major annual burden on human health, animal populations, and the global economy.1 Symptoms range from mild upper-respiratory illness to severe and fatal viral pneumonia, sepsis, and cytokine storms.2 Although the World Health Organization (WHO) updates inactivated influenza vaccines annually, influenza viruses frequently evade host immunity through antigenic shift,3 and antibodies induced by inactivated vaccines often provide limited cross-protection against divergent strains.4 Neuraminidase inhibitors can offer clinical benefit when initiated within 48 hours of symptom onset,5,6 but antiviral resistance has been increasingly reported.7 Influenza viruses are enveloped, negative-sense, single-stranded RNA viruses of the family Orthomyxoviridae and are classified into four species (IAV, IBV, ICV, and IDV). Of these, IAV causes the greatest loss of life and economic impact and is the focus of the present study.8 Because both vaccination and neuraminidase inhibition target the virus itself, and are therefore vulnerable to antigenic change and to drug resistance, strategies that instead act on the host response are an attractive complement, particularly those that strengthen cross-protective cellular immunity rather than strain-specific antibody. Dimethyl sulfoxide (DMSO) is a natural by-product of the paper industry derived from wood. It is a colorless polar solvent comprising a sulfonyl group and two methyl groups.9 Owing to its hydrophilic and lipophilic properties, DMSO is widely used in medicine.10 Clinically, DMSO is used for cryoprotective organ preservation,11,12 and the United States Food and Drug Administration (FDA) has approved DMSO for the treatment of interstitial cystitis.13 Recent studies have highlighted its therapeutic effects in autoimmune arthritis,14 pressure sores,15 complex regional pain syndrome,16 and bladder inflammation.17 Of particular relevance, DMSO has been shown to decrease herpes simplex virus (HSV) virion infectivity and halt viral replication in vitro.18 DMSO is therefore a plausible candidate for host-directed repurposing; however, its reported antiviral activity has been demonstrated in vitro, and its effect on the host immune response during a live respiratory viral infection has not been defined. What is established about the immunological activity of DMSO derives largely from non-infectious settings. We previously demonstrated that DMSO expands regulatory T (Treg) cells via signal transducer and activator of transcription 5 (STAT5) in non-obese diabetic mice.19 That observation is mechanistically relevant to antiviral immunity because STAT5 is also critical for sustaining CD8+ T cell effector responses,20 and CD8+ T cells clear influenza viruses21 and confer cross-protection against IAV and IBV by recognizing conserved viral regions;22 patients with robust CD8+ T cell responses have shorter hospital stays during severe influenza.23 CD4+ T cells, by contrast, exert dual roles: their depletion can slow viral clearance, but excessive activation may exacerbate immunopathology.24 Two questions therefore remain unanswered. First, whether systemic DMSO influences the CD8+ T cell compartment in vivo has not been examined, because our earlier work addressed Tregs in a model of autoimmunity rather than infection. Second, no in vivo study has asked whether DMSO can reshape the T cell response in the lung during respiratory viral infection, even though the lung is the compartment in which influenza is actually controlled. Given its immunomodulatory profile, we hypothesized that DMSO could be repurposed to modulate T cell-mediated immunity and improve outcomes during IAV infection. The present study therefore investigated the therapeutic efficacy of DMSO and its underlying immunological mechanisms in a murine model of H1N1 IAV infection, with emphasis on local pulmonary T cell responses and transcriptomic reprogramming. We frame this work as a hypothesis-generating study; the data presented here are correlative and any mechanistic links are presented as a working model that warrants targeted knock-out follow-up. Materials and Methods Ethics Approval All in vivo experiments were conducted in strict accordance with the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International and with the ARRIVE guidelines. The animal study protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of National Defense Medical University, under protocol number IACUC-19–204 and IACUC-21–063. All efforts were made to minimize animal suffering. Humane Endpoints and Euthanasia In strict accordance with AAALAC regulations, animal welfare was prioritized throughout the in vivo studies. Humane endpoints were established by strictly monitoring the body weight of the mice; the experimental endpoint was defined as a body weight reduction reaching 30% of the animal’s initial weight, at which point euthanasia was promptly carried out. For terminal sample collection on day 8 post-infection, euthanasia procedures were conducted in compliance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals. Mice were deeply anesthetized prior to any physical procedures using an intramuscular injection of a 1:1 mixture of Zoletil (2.5 mg/mL) and Rompun (23.32 mg/mL). Once the mice were fully anesthetized, the thoracic cavity was opened as a non-survival procedure. Depending on the downstream assay, this terminal procedure was immediately followed by either injecting 4% formalin into the trachea for lung tissue fixation or inserting a needle equipped with a PE tube into the trachea to flush the lungs with PBS for the collection of bronchoalveolar lavage fluid. Mice Five-week-old male C57BL/6 mice were purchased from the National Laboratory Animal Center (Taipei, Taiwan) and maintained until 6–8 weeks of age in the National Defense Medical University Laboratory Animal Center under specific pathogen-free, environmentally controlled conditions. Influenza Virus Propagation and Titration Recombinant influenza A/Puerto Rico/8/1934 (H1N1) was generated by reverse genetics in HEK293T cells (American Type Culture Collection [ATCC], Manassas, VA, USA) transfected with eight plasmids. After 48 hours, supernatants were propagated in Madin–Darby canine kidney (MDCK) cells. Virus stock titers were determined by plaque assay on MDCK monolayers fixed with 10% formalin and stained with 1% crystal violet. Treatment Protocol and Survival Assessment Mice were intranasally challenged with a 50% lethal dose (LD50) of A/Puerto Rico/8/1934 (H1N1). From day 1 to day 7 post-infection, mice received daily subcutaneous injections of either DMSO (0.22 g/kg/day) or an equal volume of phosphate-buffered saline (PBS) into the hindlimb. Survival was monitored daily over 21 days. Enzyme-Linked Immunosorbent Assay (ELISA) Bronchoalveolar lavage fluid (BALF) was collected and centrifuged to remove cellular debris. Cytokines were quantified using BioLegend ELISA kits according to the manufacturer’s instructions. Briefly, plates were coated with capture antibodies overnight at 4 °C, blocked for 1 hour at room temperature, and incubated with duplicate standards and samples for 2 hours. After incubation with detection antibodies (1 hour) and substrate (30 minutes), the reaction was stopped and absorbance read on a microplate reader. Cytokine concentrations (pg/mL) were interpolated from a standard curve. BALF Viral Titration On day 8 post-infection, bronchoalveolar lavage fluid (BALF) was collected as described above and clarified by centrifugation to remove cellular debris. MDCK cells were seeded in 96-well plates at 5×104 cells per well and cultured overnight, then washed with serum-free Dulbecco’s modified Eagle’s medium (DMEM) supplemented with non-essential amino acids, penicillin-streptomycin, and L-glutamine. Ten-fold serial dilutions of clarified BALF were prepared in the same medium and applied to the monolayers, which were incubated at 37 °C for 1 hour with agitation every 15 minutes to allow adsorption. The inoculum was then removed, the monolayers were washed, and serum-free DMEM containing trypsin at a 1:1000 dilution was added. After 72 hours at 37 °C the medium was removed and the monolayers were stained overnight with 0.5% methylene blue; the plates were rinsed with water and each well was scored as infected by loss of methylene blue staining, indicating cytopathic effect. The TCID50 was calculated by the Spearman-Karber method and expressed as log10 TCID50 per mL of BALF. Neuraminidase Activity Assay NA activity in bronchoalveolar lavage fluid was measured with the NA-Fluor Influenza Neuraminidase Assay Kit (Applied Biosystems/Life Technologies) according to the manufacturer’s instructions. Two-fold serial dilutions of BALF from 1:1 to 1:1024 were prepared in 1X assay buffer in black 96-well flat-bottom plates, and an equal volume of 200 µM NA-Fluor substrate working solution was added to each well. Plates were incubated at 37 °C for 60 minutes with shaking and protected from light, the reaction was terminated with NA-Fluor Stop Solution, and fluorescence was measured at an excitation wavelength of 350 to 365 nm and an emission wavelength of 440 to 460 nm. A/Puerto Rico/8/1934 (H1N1) virus stock was included as a positive control and BALF from uninfected mice as a negative control. Histological Analysis Under anesthesia, lungs were harvested, fixed in 10% formalin for 24 hours, embedded in paraffin, and sectioned. Sections were stained with hematoxylin and eosin (H&E) and examined under light microscopy. Flow Cytometry Single-cell suspensions from BALF and spleen were stained for CD45 (clone 30-F11, BioLegend), CD4 (clone GK1.5, BioLegend), CD8a (clone 53-6.7, BD), intracellular IFN-γ (clone XMG1.2, BioLegend), and a viability dye (Zombie, BioLegend). All staining antibodies were used at 1:200 in FACS buffer (CD45, CD4, CD8a) and 1:100 (intracellular IFN-γ). Surface staining was performed for 30 min at 4 °C; intracellular cytokine staining used the BD Cytofix/Cytoperm kit following 5 h ex vivo restimulation with PMA/ionomycin in the presence of brefeldin A. Data were acquired on a BD FACSLyric flow cytometer (BD Biosciences Pharmingen) and analyzed in FlowJo software. Lymphocytes were gated on forward and side scatter, followed by doublet exclusion, then viable CD45+ leukocytes; CD4+ and CD8a+ subsets were used to quantify the frequency of intracellular IFN-γ+ cells. Isolation of Splenic CD8+ T Cells Spleens were mechanically dissociated and treated with ammonium-chloride-potassium (ACK) lysis buffer to remove erythrocytes. CD8+ T cells were purified by negative selection using a Mouse CD8+ T Cell Isolation Kit according to the manufacturer’s instructions. Western Blot Splenic CD8+ T cells exposed to DMSO (0%, 0.5%, or 1.0%) for 12, 24, or 48 hours were lysed in radioimmunoprecipitation assay (RIPA) buffer containing protease inhibitors. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyvinylidene difluoride (PVDF) membranes, blocked, and incubated overnight at 4 °C with anti-LAG3 (GeneTex, GTX81605) and anti-β-actin primary antibodies. Bands were visualized by enhanced chemiluminescence (ECL) after secondary antibody incubation and quantified densitometrically. RNA Sequencing and Bioinformatic Analysis Total RNA from infected lung tissues was extracted using the RNeasy Mini Kit (Qiagen). Libraries were prepared from 300 ng RNA with TruSeq Stranded mRNA kits (Illumina). Pooled 2-nM libraries were sequenced on an Illumina NovaSeq system, yielding 20–30 million paired-end reads per sample. After quality control with fastP and multiQC, reads were aligned to the mm10 genome with HISAT2 and quantified with featureCounts. Differentially expressed genes (DEGs; adjusted p < 0.05, |log2 fold change| > 1) were identified with DESeq2 and edgeR. Functional enrichment (Gene Ontology and Kyoto Encyclopedia of Genes and Genomes [KEGG]) was performed using Enrichr (p < 0.01). Pathview visualized expression changes on KEGG pathway maps, and heatmaps were generated to compare immune-related gene sets among groups. Statistical Analysis All statistical analyses were performed using GraphPad Prism version 8.01 (GraphPad Software, San Diego, California USA). Survival curves were estimated using the Kaplan-Meier method and compared by the log-rank (Mantel-Cox) test. The hazard ratio (HR) and 95% confidence interval (CI) were calculated using the Mantel-Haenszel approach. Three-group comparisons (uninfected vs PBS vs DMSO) were performed using one-way ANOVA with Tukey’s post-hoc test; two-group comparisons used unpaired two-tailed Student’s t-test. Viral titers were log10-transformed before analysis, as is conventional for titer data spanning several orders of magnitude, and the transformed values were compared by one-way ANOVA with Tukey’s post-hoc test. Exact p-values are reported throughout. Data are presented as mean ± standard deviation (SD) unless otherwise indicated in the figure legends. A two-tailed p-value < 0.05 was considered statistically significant. Results In vivo Protective Efficacy of DMSO Against Influenza A Virus Infection To evaluate the protective efficacy of DMSO in vivo, C57BL/6 mice were intranasally challenged with an LD50 of A/PR/8/1934 (H1N1) and treated with daily subcutaneous DMSO (0.22 g/kg/day) or PBS from day 1 to day 7 post-infection (Figure 1A). Survival was monitored daily for 21 days. Kaplan-Meier analysis revealed that DMSO administration significantly reduced mortality compared with the PBS vehicle control (Figure 1B). Histopathological examination of lung tissues harvested on day 8 demonstrated extensive lymphocyte infiltration within the alveolar spaces of PBS-treated mice; in contrast, this inflammatory infiltration was attenuated in the DMSO-treated group (Figure 1C). Collectively, these results indicate that DMSO exerts a protective effect against IAV infection by improving survival and reducing the severity of viral pneumonia (21-day survival 30.7% [PBS, n = 26] vs 68% [DMSO, n = 25]; log-rank p = 0.0139; HR = 2.73, 95% CI 1.18–6.33). Figure 1In vivo protective efficacy of DMSO against influenza A virus infection. (A) Schematic representation of the experimental schedule. Mice were challenged with influenza A virus and subsequently received daily subcutaneous DMSO or PBS vehicle for seven consecutive days. (B) Kaplan-Meier survival curves of infected mice treated with DMSO (n = 25) or PBS (n = 26), monitored continuously over a 21-day observation period. Survival curves were compared by the log-rank (Mantel–Cox) test (*p < 0.05). (C) Representative hematoxylin and eosin staining of lung tissues harvested 8 days post-infection to evaluate pulmonary inflammatory infiltration.The image A shows a flowchart detailing the experimental schedule for testing DMSO′s efficacy against influenza A in C57BL/6 mice. Mice aged 5 weeks undergo adaptation until 6-8 weeks old, followed by influenza A infection (PR8, H1N1). From day 1 to day 7, mice receive daily subcutaneous injections of PBS or DMSO (0.22 g/kg). On day 21, survival rates are assessed. On day 8, samples are collected from lungs, BALF and spleen for haematoxylin and eosin staining, RNA-seq, ELISA, viral titer and flow cytometry to evaluate CD4+/CD8+ T cells and IFN-gamma expression. The image B shows a Kaplan-Meier survival graph comparing DMSO-treated mice (n=25) and PBS-treated mice (n=26) over 21 days, indicating improved survival with DMSO. The image C shows representative hematoxylin and eosin-stained lung tissue samples from uninfected control, PBS-treated and DMSO-treated mice, highlighting differences in inflammatory infiltration.Three-part image showing experimental schedule, survival graph and lung tissue samples. In vivo protective efficacy of DMSO against influenza A virus infection. (A) Schematic representation of the experimental schedule. Mice were challenged with influenza A virus and subsequently received daily subcutaneous DMSO or PBS vehicle for seven consecutive days. (B) Kaplan-Meier survival curves of infected mice treated with DMSO (n = 25) or PBS (n = 26), monitored continuously over a 21-day observation period. Survival curves were compared by the log-rank (Mantel–Cox) test (*p < 0.05). (C) Representative hematoxylin and eosin staining of lung tissues harvested 8 days post-infection to evaluate pulmonary inflammatory infiltration. DMSO Treatment is Associated with a Lower Day-8 Viral Burden in the Airway To determine whether the survival benefit reflected reduced viral replication, preserved tissue integrity at a comparable viral burden, or both, BALF collected on day 8 post-infection was analysed in parallel by neuraminidase (NA) activity assay and TCID50 titration (Figure 2A). NA activity was readily detectable in BALF from PBS-treated mice across the dilution series, whereas activity in DMSO-treated mice was lower and approached that of uninfected controls (Figure 2B). Consistent with this, infectious virus titers were significantly lower in DMSO-treated than in PBS-treated mice (mean titers 2.2×103 vs 4.3×104 TCID50/mL, n = 3 per group; p = 0.034 by one-way ANOVA with Tukey’s post-hoc test on log10-transformed titers; Figure 2C–E). These results indicate that DMSO lowers the infectious viral burden in the airway at the same timepoint at which the histological, cytokine, flow-cytometric, and transcriptomic analyses were performed, and therefore that the improved outcome reflects a combination of reduced viral burden and modulation of the host response. Figure 2Day-8 bronchoalveolar lavage fluid (BALF) viral burden assessed by neuraminidase activity assay and TCID50 titration. (A) Schematic of the experimental workflow. BALF was collected on day 8 post-infection from uninfected (un-If.), PBS-treated, and DMSO-treated mice and analysed in parallel by neuraminidase (NA) activity assay and TCID50 assay. (B) NA activity in BALF across two-fold dilutions (1:1 to 1:1024), expressed as relative fluorescence units. A/Puerto Rico/8/1934 (H1N1) virus stock served as the positive control (n = 3 per group). (C) Infectious virus titers in BALF determined as TCID50 and expressed as log10 TCID50/mL (n = 3 mice per group). Bars show the mean and error bars the standard error of the mean. Statistical significance was determined using one-way ANOVA with Tukey post-hoc test on log10-transformed titers (*p < 0.05). (D and E) Representative methylene blue-stained TCID50 assay plates from PBS-treated (D) and DMSO-treated (E) mice.Image A illustrates a workflow: un-infected, PBS-treated and DMSO-treated mice under infection treatment. Day 8 BALF collection leads to NA activity and TCID50 assays. Image B presents a line graph of NA activity. X-axis: Virus dilution (1:1 to 1:1024). Y-axis: Relative fluorescence unit (0 to 12000) with an axis break(lower segment 0–200, upper segment 3000–12000). Curves: H1N1, un-infected, PBS, DMSO (n=3 each). H1N1 starts at 12000, declines to 3000. PBS starts at 150, declines to 80. DMSO and un-infected remain around 70-80. Image C shows a bar chart of virus titers. Y-axis: log TCID50/ml (10 superscript 0 to 10 superscript 5). X-axis: un-infected, PBS, DMSO (n=3 each). PBS near 10 superscript 4, DMSO near 10 superscript 3, with asterisk indicating difference. Image D: TCID50 assay plate labeled PBS. Image E: TCID50 assay plate labeled DMSO.A multi-panel infographic of BALF workflow, NA dilution curve, TCID50 bar chart and assay plates. Day-8 bronchoalveolar lavage fluid (BALF) viral burden assessed by neuraminidase activity assay and TCID50 titration. (A) Schematic of the experimental workflow. BALF was collected on day 8 post-infection from uninfected (un-If.), PBS-treated, and DMSO-treated mice and analysed in parallel by neuraminidase (NA) activity assay and TCID50 assay. (B) NA activity in BALF across two-fold dilutions (1:1 to 1:1024), expressed as relative fluorescence units. A/Puerto Rico/8/1934 (H1N1) virus stock served as the positive control (n = 3 per group). (C) Infectious virus titers in BALF determined as TCID50 and expressed as log10 TCID50/mL (n = 3 mice per group). Bars show the mean and error bars the standard error of the mean. Statistical significance was determined using one-way ANOVA with Tukey post-hoc test on log10-transformed titers (*p < 0.05). (D and E) Representative methylene blue-stained TCID50 assay plates from PBS-treated (D) and DMSO-treated (E) mice. DMSO is Associated with a Modulated Local Pro- and Anti-Inflammatory Cytokine Profile in the Lungs Having established the overall protective effect of DMSO, we next investigated its impact on the local pulmonary inflammatory milieu. Cytokines were quantified using individual sandwich ELISA kits on day 8 BALF (Figure 3A). The concentration of interferon-gamma (IFN-γ) was significantly elevated in the DMSO-treated group compared with both uninfected and PBS-treated controls (Figure 3B). In contrast, the pro-inflammatory cytokines TNF-α, IL-6, and IL-1β were not further increased by DMSO relative to the PBS vehicle (Figure 3C–E). DMSO treatment also robustly increased levels of the anti-inflammatory and tissue-repair cytokine transforming growth factor-beta (TGF-β; Figure 3G), without further altering IL-10 (Figure 3F). These data suggest that DMSO intensifies the localized antiviral response primarily through selective enhancement of IFN-γ accompanied by pro-resolving signals. Figure 3ELISA quantification of pro- and anti-inflammatory cytokines in bronchoalveolar lavage fluid (BALF). (A) Schematic representation of the experimental workflow for day-8 BALF collection, processing, and sandwich ELISA analysis. (B–G) Concentrations (pg/mL) of local cytokines, including (B) IFN-γ, (C) TNF-α, (D) IL-6, (E) IL-1β, (F) IL-10, and (G) TGF-β, were measured in BALF of uninfected mice (un-If.), infected mice treated with PBS vehicle, and infected mice treated with DMSO. Data are expressed as mean ± SD (n = 3 mice per group). Statistical significance was determined using one-way ANOVA with Tukey post-hoc (*p < 0.05; **p < 0.01).The schematic outlines the day-8 bronchoalveolar lavage fluid (BALF) collection and analysis workflow. Mice are divided into uninfected, PBS-treated and DMSO-treated groups. The procedure includes infection treatment, BALF collection, centrifugation, supernatant collection and ELISA. Bar graphs show cytokine concentrations in pg/mL. Graph B: IFN-gamma is highest in DMSO-treated mice. Graph C: TNF-alpha levels are similar in PBS and DMSO treatments. Graph D: IL-6 levels are similar in PBS and DMSO treatments. Graph E: IL-1beta is highest in uninfected mice. Graph F: IL-10 is highest in uninfected mice. Graph G: TGF-beta is highest in DMSO-treated mice. Statistical significance is marked by asterisks: one for p < 0.05, two for p < 0.01.A schematic and bar graphs of cytokine levels in BALF from treated mice. ELISA quantification of pro- and anti-inflammatory cytokines in bronchoalveolar lavage fluid (BALF). (A) Schematic representation of the experimental workflow for day-8 BALF collection, processing, and sandwich ELISA analysis. (B–G) Concentrations (pg/mL) of local cytokines, including (B) IFN-γ, (C) TNF-α, (D) IL-6, (E) IL-1β, (F) IL-10, and (G) TGF-β, were measured in BALF of uninfected mice (un-If.), infected mice treated with PBS vehicle, and infected mice treated with DMSO. Data are expressed as mean ± SD (n = 3 mice per group). Statistical significance was determined using one-way ANOVA with Tukey post-hoc (*p < 0.05; **p < 0.01). DMSO Is Selectively Associated with a Higher Frequency of IFN-γ-Producing CD8+ T Cells in the Pulmonary Microenvironment To identify the cellular sources of the modulated cytokine response, we used flow cytometry to enumerate IFN-γ-producing CD4+ and CD8+ T cells in both the peripheral spleen and local BALF (Figure 4A and B). Consistent with a localized rather than systemic mechanism, no significant differences in IFN-γ+ CD4+ or CD8+ frequencies were observed in the spleens of DMSO-treated mice compared with PBS controls (Figure 4C and D). Within BALF, the proportion of IFN-γ+ CD4+ T cells also remained unchanged (Figure 4E). However, DMSO treatment was associated with a statistically significant increase in the frequency of IFN-γ-producing CD8+ T cells in the lungs (Figure 4F). Together, these findings indicate that systemic DMSO is selectively associated with a reshaped local pulmonary inflammatory environment by enriching activated, antiviral CD8+ T cells at the primary site of infection. Figure 4Flow cytometric analysis of IFN-γ–producing T cell populations in the spleen and lungs. (A) Experimental pipeline for the isolation and flow cytometric gating of single-cell suspensions derived from the spleen and BALF on day 8 post-infection (n = 3 mice per group). (B) Representative flow cytometric gating strategy for identification of IFN-γ–producing T cells: lymphocytes were gated on forward and side scatter (FSC vs SSC), followed by doublet exclusion (FSC-A vs FSC-H). Viable leukocytes were identified as CD45+ and Zombie-negative; CD4+ and CD8a+ T cell subsets were then gated to quantify intracellular IFN-γ. (C and D) Representative dot plots and quantitative bar graphs showing the frequency of IFN-γ+ cells within CD4+ (C) and CD8+ (D) T cell compartments in the spleen. (E and F) Representative dot plots and quantitative bar graphs showing the frequency of IFN-γ+ cells within CD4+ (E) and CD8+ (F) T cell compartments in BALF (*p < 0.05).The image A showing a flowchart of a workflow from spleen and bronchoalveolar lavage fluid collection to ex vivo stimulation, fluorophore staining, flow cytometer measurement and a branching data gating strategy. The image B showing a sequence of flow cytometry dot plots for gating. Plot 1 has x axis labeled, FSC and y axis labeled, SSC. A lymphocyte gate is drawn. Plot 2 has x axis labeled, FSC A and y axis labeled, FSC H, with a single cell gate. Plot 3 has x axis labeled, Zombie dye and y axis labeled, CD45, with a viable leukocyte gate. Plot 4 has x axis labeled, CD4 and y axis labeled, FSC, with a CD4 T cell gate. Plot 5 has x axis labeled, CD8a and y axis labeled, FSC, with a CD8 T cell gate. Plot 6 has x axis labeled, CD4 or CD8a and y axis labeled, interferon gamma, with a gate labeled percent of interferon gamma. The image C showing two dot plots and a bar chart for CD4. Dot plots have x axis labeled, CD4 APC and y axis labeled, interferon gamma FITC. The top dot plot labeled PBS shows 3.99 percent. The bottom dot plot labeled dimethyl sulfoxide shows 3.92 percent. The bar chart x axis shows PBS and dimethyl sulfoxide. The y axis is labeled, percent of interferon gamma positive cells within CD4+ T cells, with ticks at 0 and 5. Bars are near 2 for PBS and near 3 for dimethyl sulfoxide. The image D showing two dot plots and a bar chart for CD8. Dot plots have x axis labeled, CD8 PerCP Cy5.5 and y axis labeled, interferon gamma FITC. The top dot plot labeled PBS shows 8.74 percent. The bottom dot plot labeled dimethyl sulfoxide shows 10.7 percent. The bar chart x axis shows PBS and dimethyl sulfoxide. The y axis is labeled, percent of interferon gamma positive cells within CD8+ T cells, with ticks at 0, 2, 4, 6 and 8. Bars are near 5 for PBS and near 5.2 for dimethyl sulfoxide. The image E showing two dot plots and a bar chart for CD4 in bronchoalveolar lavage fluid. Dot plots have x axis labeled, CD4 APC and y axis labeled, interferon gammaFITC. The top dot plot labeled PBS shows 42.0 percent. The bottom dot plot labeled dimethyl sulfoxide shows 51.5 percent. The bar chart x axis shows PBS and dimethyl sulfoxide. The y axis is labeled, percent of interferon gamma positive cells within CD4+ T cells, with ticks at 0, 10, 20, 30, 40 and 50. Bars are near 18 for PBS and near 32 for dimethyl sulfoxide. The image F showing two dot plots and a bar chart for CD8 in bronchoalveolar lavage fluid. Dot plots have x axis labeled, CD8 PerCP Cy5.5 and y axis labeled, interferon gamma FITC. The top dot plot labeled PBS shows 37.9 percent. The bottom dot plot labeled dimethyl sulfoxide shows 53.6 percent. The bar chart x axis shows PBS and dimethyl sulfoxide. The y axis is labeled, percent of interferon gamma positive cells within CD8+ T cells, with ticks at 0, 20, 40, 60 and 80. Bars are near 28 for PBS and near 55 for dimethyl sulfoxide, with an asterisk above a bracket spanning the two bars.Flow cytometry dot plots and bar charts of interferon gamma positive T cells across groups. Flow cytometric analysis of IFN-γ–producing T cell populations in the spleen and lungs. (A) Experimental pipeline for the isolation and flow cytometric gating of single-cell suspensions derived from the spleen and BALF on day 8 post-infection (n = 3 mice per group). (B) Representative flow cytometric gating strategy for identification of IFN-γ–producing T cells: lymphocytes were gated on forward and side scatter (FSC vs SSC), followed by doublet exclusion (FSC-A vs FSC-H). Viable leukocytes were identified as CD45+ and Zombie-negative; CD4+ and CD8a+ T cell subsets were then gated to quantify intracellular IFN-γ. (C and D) Representative dot plots and quantitative bar graphs showing the frequency of IFN-γ+ cells within CD4+ (C) and CD8+ (D) T cell compartments in the spleen. (E and F) Representative dot plots and quantitative bar graphs showing the frequency of IFN-γ+ cells within CD4+ (E) and CD8+ (F) T cell compartments in BALF (*p < 0.05). DMSO is Associated with Transcriptional Changes in Immune-Related Genes in the Lungs To elucidate the intrinsic molecular mechanisms underlying DMSO’s immunomodulatory effects, we performed RNA sequencing on whole-lung tissues (Figure 5A). Differentially expressed gene (DEG) analysis identified a targeted subset of transcriptional alterations induced by DMSO compared with the PBS control. As shown in the volcano plot (Figure 5B), significantly upregulated transcripts in the DMSO cohort included Ms4a7, Rpl15, Mmp12, and Igf1, which are predominantly associated with tissue remodeling, cellular growth, and immune regulation. KEGG pathway enrichment analysis further showed that these DEGs were concentrated in immune and signal-transduction cascades, identifying “cytokine–cytokine receptor interaction” and the “PI3K-Akt signaling pathway” among the most enriched networks (Figure 5C, Supplementary Figure 1). These signaling axes are essential for leukocyte activation, survival, and directed migration. Figure 5Transcriptomic profiling of whole lung tissues by RNA sequencing. (A) Methodological workflow depicting RNA extraction, mRNA enrichment, library preparation, and high-throughput Illumina NovaSeq sequencing. (B) Volcano plot of differentially expressed genes (DEGs) between DMSO-treated (n = 3) and PBS-treated (n = 3) groups, plotted as log2 fold change against -log10 adjusted p value (padj). Red, significantly upregulated genes; blue, significantly downregulated genes; black, no significant change. Dashed lines indicate the DEG thresholds (padj < 0.05 and |log2 fold change| > 1). Triangles at the upper margin denote genes whose padj fell below the smallest value displayed on the axis. (C) Bar chart of KEGG pathway enrichment analysis displaying the top statistically significant immune, metabolic, and signal-transduction pathways modulated by DMSO.The image A showing a left to right workflow diagram with icons of lungs and a mouse, followed by lab tubes and a sequencing machine. Text under each step reads, Mouse lung tissue, Total RNA Extraction, mRNA Enrichment left parenthesis Poly-A selection right parenthesis, Fragmentation and cDNA Synthesis, Library Prep, Illumina NovaSeq. The image B showing a scatter volcano plot titled, DEGseq Volcano. The x-axis label reads, log2FoldChange. The y-axis label reads, minus log10 left parenthesis padj right parenthesis. Points form a dense vertical cluster near x equals 0, with points extending left and right. A small key at right lists, Up, Down, NoDiff. Several points are annotated with gene labels, including, Igf1, Rpl15, Mmp12. The image C showing a horizontal bar chart. The x-axis label reads, Count. The y-axis lists pathway names including, ECM-receptor interaction, Cytokine-cytokine receptor interaction, Focal adhesion, Malaria, PI3K-Akt signaling pathway, Cell adhesion molecules, Calcium signaling pathway, Cytoskeleton in muscle cells, Hippo signaling pathway, Human papillomavirus infection. A vertical color scale at right is labeled, p.adjust, with tick labels 0.001, 0.002, 0.003, 0.004 and 0.005. Bars span roughly from about 15 to about 40 on the Count axis. Standard scientific plots with no decorative styling.Three-part scientific figure of mouse lung RNA sequencing workflow, DEGseq volcano plot and KEGG bar chart. Transcriptomic profiling of whole lung tissues by RNA sequencing. (A) Methodological workflow depicting RNA extraction, mRNA enrichment, library preparation, and high-throughput Illumina NovaSeq sequencing. (B) Volcano plot of differentially expressed genes (DEGs) between DMSO-treated (n = 3) and PBS-treated (n = 3) groups, plotted as log2 fold change against -log10 adjusted p value (padj). Red, significantly upregulated genes; blue, significantly downregulated genes; black, no significant change. Dashed lines indicate the DEG thresholds (padj < 0.05 and |log2 fold change| > 1). Triangles at the upper margin denote genes whose padj fell below the smallest value displayed on the axis. (C) Bar chart of KEGG pathway enrichment analysis displaying the top statistically significant immune, metabolic, and signal-transduction pathways modulated by DMSO. DMSO is Associated with Accelerated in vitro LAG3 Kinetics in CD8+ T Cells and Maintains Physiological Homeostasis To verify the accelerated CD8+ T cell activation kinetics suggested by the transcriptomic data, an in vitro stimulation assay was performed using isolated CD8+ splenocytes exposed to varying concentrations of DMSO (Figure 6A). Immunoblot analysis revealed that DMSO (0.5% and 1.0%) accelerated the onset of LAG3 expression at 12, 24, and 48 hours compared with untreated controls (Figure 6B). Because LAG3 upregulation typically follows robust T cell activation as a homeostatic immune checkpoint, these dynamics are consistent with earlier CD8+ T cell activation in the presence of DMSO, accompanied by physiological feedback that limits overstimulation. Figure 6Accelerated in vitro expression of LAG3 in DMSO-treated CD8+ T cells. (A) Schematic of the in vitro experimental design. Splenic CD8+ T cells were isolated and stimulated in the presence of 0%, 0.5%, or 1.0% DMSO. Cell lysates were collected at 12, 24, and 48 hours for protein extraction. (B) Representative Western blot of LAG3 and β-actin (loading control). The numerical values below the LAG3 bands indicate the relative densitometric quantification (ratio of LAG3 to β-actin) at each respective time point and condition.The image consists of two parts. A shows a schematic of the in vitro experimental design. Mouse spleen-derived CD8+ T cells are isolated and treated with 0 percent, 0.5 percent, or 1.0 percent DMSO. Cells are collected at 12, 24 and 48 hours for cell lysis and protein extraction, followed by gel electrophoresis and transfer for detection via Western blot. B shows a Western blot analysis of LAG3 and beta-actin as a loading control. The blot displays bands for LAG3 at 12, 24 and 48 hours under different DMSO concentrations. Below the LAG3 bands, numerical values indicate the relative densitometric quantification ratio of LAG3 to beta-actin: 1.0, 1.74, 1.77 at 12 hours; 1.49, 1.56, 2.47 at 24 hours; and 1.94, 2.35, 1.85 at 48 hours.In vitro expression of LAG3 in DMSO-treated CD8+ T cells with schematic and Western blot analysis. Accelerated in vitro expression of LAG3 in DMSO-treated CD8+ T cells. (A) Schematic of the in vitro experimental design. Splenic CD8+ T cells were isolated and stimulated in the presence of 0%, 0.5%, or 1.0% DMSO. Cell lysates were collected at 12, 24, and 48 hours for protein extraction. (B) Representative Western blot of LAG3 and β-actin (loading control). The numerical values below the LAG3 bands indicate the relative densitometric quantification (ratio of LAG3 to β-actin) at each respective time point and condition. DMSO is Associated with Distinct Regulatory Gene Signatures in the Lung Comparative heatmap analysis (Figure 7) demonstrated that DMSO treatment is associated with a refined pulmonary transcriptomic landscape involving specific regulatory modules. Within the mitogen-activated protein kinase (MAPK) signaling axis, DMSO was associated with selective modulation of MAPK-family members and notably enhanced expression of Mapk4, Mapk11, and Mapk12 (Figure 7A), providing a potential signal for cellular adaptation. Although T-bet (Tbx21) expression was primarily driven by viral challenge, DMSO independently upregulated a distinct cluster of T-box transcription factors, including Tbx1 and Tbx4 (Figure 7B), accompanied by induction of the immune-modulatory factors Foxp3. DMSO-treated tissues also exhibited increased expression of metabolic and cell-cycle regulators, particularly Nek1, Nek7, and Pparg (Figure 7C). Together, these transcript-level associations are compatible with a shift in the lung microenvironment in which selective T-box modulation and a metabolic priming signature accompany the reinforced CD8+ T cell axis. Figure 7Transcriptional heatmaps of differential expression profiles across key immune and metabolic regulatory families. Heatmaps display the log2 fold-change (log2FC) of specific gene clusters across three comparative groups: PBS vs uninfected (un-if), DMSO vs un-if, and DMSO vs PBS (n = 3 mice per group). Red, relative upregulation; blue, downregulation. (A) Expression alterations within the mitogen-activated protein kinase (MAPK) signaling cascade. (B) Expression profiles of master transcriptional regulators, including the T-box (Tbx) family, GATA family, and Foxp3. (C) Differential expression of metabolic regulators (Pparg) and cell-cycle–associated NIMA-related kinases (Nek family).Image A showing a heatmap of Mapk family genes with columns labeled PBS vs un-if, DMSO vs un-if and DMSO vs PBS. Rows list Mapk7 through Mapk15. A Log2FC legend ranges from minus 5 to 4, where higher positive Log2FC indicates upregulation, lower negative Log2FC indicates downregulation and values near 0 indicate little change. Visible pattern: Mapk12 is strongly downregulated in PBS vs un-if and DMSO vs un-if, while Mapk10 is strongly upregulated in PBS vs un-if; many other Mapk rows are near 0 to moderately positive in the first two comparisons and closer to 0 in DMSO vs PBS. Image B showing a heatmap of transcriptional regulators with the same three columns. Rows include Rorc, Rora, Prorsd1, Ror1, multiple Tbx genes, multiple Gata genes and Foxp3. The Log2FC legend ranges from minus 4 to 5 with the same meaning for positive, negative and near 0 values. Visible pattern: Tbx21 shows strong upregulation in PBS vs un-if and DMSO vs un-if, while Rorc and Tbx1 show strong downregulation in PBS vs un-if; several Gata genes are moderately upregulated in the first two comparisons and closer to 0 in DMSO vs PBS. Image C showing a heatmap of Nek genes and metabolic regulators with the same three columns. Rows include Nek8, Nek4, Nek7, Nek2, Nek6, Nek3, Nek1, Nek9, Nek11, Nek5, Nek10, Pparg, Ppard, Ppargc1a, Ppargc1b and Nrf1. The Log2FC legend ranges from minus 3 to 2 with the same meaning for positive, negative and near 0 values. Visible pattern: Nek2 and Ppargc1a are strongly upregulated in PBS vs un-if and DMSO vs un-if, while Nek11 and Pparg are strongly downregulated in PBS vs un-if and DMSO vs un-if; DMSO vs PBS is often nearer 0 for many rows, with some rows shifting to moderate upregulation.Three heatmaps of log2 fold-change for Mapk, transcription factors and Nek and Pparg genes across comparisons. Transcriptional heatmaps of differential expression profiles across key immune and metabolic regulatory families. Heatmaps display the log2 fold-change (log2FC) of specific gene clusters across three comparative groups: PBS vs uninfected (un-if), DMSO vs un-if, and DMSO vs PBS (n = 3 mice per group). Red, relative upregulation; blue, downregulation. (A) Expression alterations within the mitogen-activated protein kinase (MAPK) signaling cascade. (B) Expression profiles of master transcriptional regulators, including the T-box (Tbx) family, GATA family, and Foxp3. (C) Differential expression of metabolic regulators (Pparg) and cell-cycle–associated NIMA-related kinases (Nek family). Discussion Interferons are cytokines with broad antiviral and immunoregulatory activity, classified into three types according to their receptors.25 Type II interferon, IFN-γ, derived from CD8+ T cells, natural killer (NK) cells, and macrophages, plays a pivotal role in antiviral defense. IFN-γ enhances CD8+ T cell effector function and differentiation, augments antigen processing and presentation via major histocompatibility complex (MHC) class I,26 and stimulates macrophage nitric oxide production and phagocytic activity.27 In the present study, the frequency of IFN-γ–producing CD8+ T cells was significantly elevated in BALF on day 8 of DMSO treatment, indicating an enhanced antiviral response. Critically, this enhancement was confined to BALF and was not detected in the spleen, supporting a localized rather than systemic mechanism of DMSO action. Cytotoxic CD8+ T cells eliminate virus-infected cells, secrete antiviral cytokines, and confer broadly protective immunity against influenza viruses. Following IAV infection, pulmonary dendritic cells license CD8+ T cell proliferation and activation via MHC I; activated CD8+ T cells then secrete perforin, granzyme, and tumor necrosis factor, triggering caspase-mediated apoptosis of infected targets.28 We observed that DMSO increased CD8+ T cell activation status, supporting CD8+ T cells as a plausible effector mechanism for DMSO. Although LAG3 is canonically characterized as an inhibitory receptor associated with T cell exhaustion, its rapid upregulation following initial stimulation is also a hallmark of vigorous T cell activation. The accelerated LAG3 induction observed in vitro is consistent with a priming effect of DMSO on CD8+ T cells. These correlative data are compatible with a model in which DMSO accelerates early CD8+ T cell activation while engaging physiological feedback that limits overstimulation; formal proof of causality will require CD8+ T cell depletion or genetic loss-of-function studies. This balance likely contributes to the observed reduction in pulmonary inflammatory infiltration and the suppression of neutrophil-recruiting chemokines.29 LAG3 upregulation is a dual marker reflecting both vigorous activation and inhibitory feedback; our in vitro data alone cannot distinguish between these states. Future work using granzyme B, perforin, and CD69 co-staining and, ideally, single-cell RNA-seq at matched time points, will be required to clarify whether DMSO-treated LAG3+ CD8+ T cells in our system are activated effectors, terminally exhausted cells, or both. The present study also asked whether this reinforced CD8+ T cell response is accompanied by control of the virus itself. BALF viral titers measured on day 8, the same timepoint as the immunological analyses, were lower in DMSO-treated mice, and neuraminidase activity in BALF was correspondingly reduced. The improved outcome is therefore not explained by tolerance of an equivalent viral burden. Two mechanisms could contribute, and the present data are most consistent with a combination of both. First, an accelerated and locally enriched IFN-γ–producing CD8+ T cell response would be expected to clear infected cells more efficiently, which links the immunological changes described above to the lower viral burden. Second, DMSO restricts herpes simplex virus replication in vitro,18 so a direct effect on viral replication cannot be excluded. Because CD8+ T cell depletion and loss-of-function experiments were not performed, the relative contribution of immune-mediated clearance and of direct antiviral restriction cannot be resolved here; the reduced viral burden and the reduced immunopathology are therefore best presented as co-occurring features of the DMSO response rather than as a single causal chain. Overactive induction of pro-inflammatory cytokines, coupled with inadequate anti-inflammatory regulation, is the hallmark of cytokine storm during influenza infection and is a major contributor to multi-organ dysfunction and mortality.30 The role of anti-inflammatory cytokines is to suppress uncontrolled inflammation and to preserve immune homeostasis, and Treg activation has been shown to prevent cytokine storm in animal models.31 Our previous research demonstrated that DMSO acts as an immunosuppressive agent by expanding Tregs through STAT5 signaling and inhibiting T helper 1 responses in non-obese diabetic mice;19 however, STAT signaling has a context-dependent dual role in immunology. Tripathi et al showed that STAT5 is indispensable for the maintenance of effector CD8+ T cell responses.20 Our findings are compatible with a model in which DMSO acts as an immunomodulator associated with a reinforced CD8+ T cell axis, without a concomitant rise in the pro-inflammatory cytokines TNF-α, IL-6, or IL-1β. By coordinating an extracellular cytokine milieu enriched in IL-2, IL-4, and IL-2132,33 with the intracellular upregulation of Nek kinases and PI3K/Akt signaling, DMSO appears to create a permissive environment for T cell proliferation and signal transduction.34,35 The concurrent induction of T-box transcription factors36 and IL-1837 likely reinforces the cytotoxic machinery required for viral clearance. Importantly, the upregulation of Pparg and Foxp3, together with selective suppression of neutrophil-recruiting CXCL5/6/7 chemokines, highlights an additional regulatory layer.38 This dual-action mechanism suggests that DMSO does not merely amplify the immune response; rather, it refines it, reinforcing targeted CD8+ effector functions and metabolic fitness while mitigating the immunopathology and collateral tissue damage typical of acute viral infection. In contrast to CD8+ T cells, our data indicate that CD4+ T cells are not significantly affected by DMSO treatment. Although previous in vitro work reported that DMSO can inhibit CD4+ T cell proliferation,39 we did not observe a reduction in CD4+ T cell numbers in vivo. Flow cytometry showed no increase in IFN-γ–producing CD4+ T cells in either the spleen or BALF. Together, these results indicate that the DMSO-associated changes were selective for CD8+ T cells, with minimal or modestly suppressive effects on CD4+ T cells. DMSO is an established penetration enhancer that increases membrane fluidity and, at higher concentrations, forms transient water pores, thereby improving the solubility, permeability, and bioavailability of co-administered compounds.13 Its effects could therefore in principle reflect enhanced delivery of another molecule rather than an intrinsic action. In this study, however, DMSO was given as the sole agent, with no co-formulated or co-administered drug and with PBS as the comparator, so no second compound was available for enhanced delivery. An intrinsic immunomodulatory action of DMSO administered alone is moreover well documented, in the repression of inflammatory cytokine production by human blood cells,14 the suppression of bladder inflammation in vivo,17 the restriction of herpes simplex virus replication in vitro,18 and the STAT5-dependent expansion of regulatory T cells in non-obese diabetic mice.19 The present findings extend this intrinsic activity to the pulmonary CD8+ T cell response during influenza A virus infection. We cannot exclude that the membrane-permeabilising property of DMSO alters the distribution or handling of endogenous solutes, which warrants further study. This study has several strengths. To our knowledge, it is the first investigation to evaluate the impact of DMSO on IAV infection in an in vivo animal model, addressing a gap in the literature. By integrating survival monitoring, histopathological assessment, viral titration and neuraminidase activity assay, sandwich ELISA cytokine profiling, flow cytometry, and high-throughput RNA sequencing, we provide a comprehensive overview of how DMSO fine-tunes the CD8+ T cell axis through modulation of cell-cycle, metabolic, and effector regulators, including Nek kinases and the PI3K/Akt signaling axis. The precise identification of DMSO’s localized effect within the lung, rather than the systemic splenic environment, also offers important insights into its selective immunomodulatory mechanism. Several limitations warrant consideration. (i) BALF viral titers were measured on day 8 post-infection and were significantly lower in DMSO-treated mice, indicating that the improved outcome is not attributable to modulation of the host response at an equivalent viral burden alone; the present data cannot, however, separate a direct antiviral effect of DMSO from accelerated immune-mediated viral clearance. (ii) CD8 antibody depletion and Cd8a–/– or Ifng–/– genetic studies were not performed; the causal contribution of CD8+ T cells to the survival benefit cannot be formally inferred from the present data and is presented as a working model only. (iii) Only a single viral strain (A/PR/8/34, H1N1) was tested, so cross-protection against H3N2 / H5N1 strains remains an open question. (iv) Bulk RNA-seq cannot resolve which cell types drive the pathway-level signals; single-cell RNA-seq is planned. (v) The study was conducted exclusively in a C57BL/6 mouse model; immunological differences between mice and humans necessitate caution when extrapolating to clinical settings. (vi) Although we observed enhanced IFN-γ secretion and augmented CD8+ T cell activity, the precise intracellular molecular triggers and direct receptor-binding targets of DMSO remain to be fully elucidated. (vii) The in vitro LAG3 kinetic assay used splenic CD8+ T cells, whereas the DMSO-associated expansion in vivo was confined to the lung; extrapolating this in vitro splenic observation to a pulmonary phenomenon is a limitation, and confirmation in lung- or airway-derived CD8+ T cells is warranted. (viii) The in vitro DMSO concentrations (0.5% and 1.0%, approximately 70 and 140 mM) exceed systemic exposure in vivo, where plasma DMSO reaches only the low-millimolar range;40 the in vitro data should therefore be read as a concentration-amplified mechanistic probe rather than a model of the administered dose, and the response at physiologically achievable concentrations requires a dedicated dose-response study. (ix) Several mechanistic experiments (BALF ELISA, flow cytometry, and whole-lung RNA sequencing) were performed with n = 3 mice per group. This small sample size limits statistical precision and power and may estimate effect sizes imprecisely; the corresponding results should be regarded as exploratory and require confirmation in adequately powered cohorts. Conclusion In summary, daily subcutaneous DMSO is associated with improved survival and reduced pulmonary inflammation in a murine A/PR/8/34 (H1N1) IAV model. At day-8, viral burden in the airway was also lower in DMSO-treated mice, indicating that the benefit reflects a combination of reduced viral burden and modulation of the host response. This protective benefit co-occurs with an increased frequency of pulmonary IFN-γ+ CD8+ T cells, alongside the up-regulation of PI3K/Akt and T-box transcriptional programs. Ultimately, these correlative findings support follow-up CD8 depletion or knock-out experiments to test causality, and motivate the evaluation of DMSO as a host-directed adjunct therapy for influenza.