This study was conducted according to the guidelines outlined by the National Science Foundation Animal Welfare Requirements and the Wakayama Medical University Animal Care and Use Committee. The study was approved by the Institutional Animal Care and Use Committee at Wakayama Medical University (Approval number: 901).

A pneumococcal strain, P2431 (clinical isolate of serotype 6A), which can establish stable colonization and cause sepsis, was used in this study (18). The bacteria were grown in Todd–Hewitt broth containing 0.5% yeast extract (THY) at 37°C until the mid-log phase and stocked in aliquots at known CFU concentrations in THY broth containing 10% glycerol at –80°C until the experiment.

The mice used in this experiment were around 6 weeks old. C57BL/6J wild-type mice were purchased from Charles River Laboratories Japan, Inc. (Yokohama, Japan). TRPV1 KO mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA), and TRPV4 KO mice were introduced from RIKEN BRC (Tsukuba, Japan) (19, 20). All mice were maintained in a conventional animal facility at Wakayama Medical University.

The schematic of the experiments is shown in Figure 1 . This study was designed to investigate a role of TRPV channels in the course of the development of invasive pneumococcal infections following nasal colonization. For the pneumococcal mono-infection group, the mice were intranasally administered with 10 µl of phosphate-buffered saline (PBS) on day –2, and with S. pneumoniae on day 0. For the influenza A virus (IAV) coinfection group, the mice were intranasally administered with 10 µl of IAV (HKx31, H3N2 strain) (2 × 10³ TCID₅₀) on day –2, and with S. pneumoniae on day 0. To avoid the mice aspirating the bacteria to the lower respiratory tract and causing pneumonia directly, the mice were inoculated to colonize S. pneumoniae intranasally without anesthesia, as previously reported (21–23). The mice were sacrificed by isoflurane asphyxiation for sampling on the desired experimental day or monitored until the mice showed signs of lethal infection including sepsis and sustained bacteremia (decreased motor activity, shivering, and messy hair).

The upper respiratory tract was lavaged with 200 µl of sterile PBS from a needle (26 gauge) inserted into the trachea, and the fluid was collected in a sterile tube. Twenty microliters of the lavage was serially diluted and incubated on a blood plate overnight so the colonies could be counted. Approximately 500 µl of blood was collected by cardiac puncture from each mouse. After taking 20 µl for colony counting, 480 µl of the blood was centrifuged and the serum was stocked for ELISA. The lungs, spleens, hearts, and brains were obtained after the removal of the blood by cardiac puncture. After washing thoroughly with sterile PBS, the organs were homogenized in 1 ml of PBS for counting colonies or were put in 4% of paraformaldehyde (PFA) for histology.

The spleens and the lungs were placed on the tubes consisting of 1 ml PFA. The organs were then placed in the automatic sample processor and dehydrated for 48 h. After the cooling process, the tissues were each cut 5 mm thick by microtome and were stored until hematoxylin and eosin (H & E) staining analysis.

The paraffin was fixed for 20 min in the incubator (67°C). The slides were cooled at room temperature for 15 min. The sections were deparaffinized with lemosol two times for 5 min each. The sections were then treated in the following order: 100% ethanol for 1 min × 2, 95% ethanol for 1 min × 2, 80% ethanol for 1 min, 70% ethanol for 1 min. The slides were soaked in Mayer’s hematoxylin solution for 7 min and then rinsed in tap water. The slides were soaked in an eosin alcohol solution for 1.5 min and then rinsed in tap water. The tissues were dehydrated in the following order for 1 min each: 70%, 80%, 95%, 95%, 100%, and 100% ethanol). The tissues were cleaned in lemosol for 1 min two times. Then, the slides were coverslipped with Canada balsam mounting solution.

The severity of lung inflammation was quantified using the scoring system proposed in a previous report (24). In brief, the lung tissue was evaluated by five independent factors (neutrophils in the alveolar space, neutrophils in the interstitial space, hyaline membranes, proteinaceous debris filling the air spaces, alveolar septal thickening) indicating the degree of inflammation and calculated using the original formula ( Supplementary Table ). Six samples were evaluated for each group.

Nasal lavages were pelleted at 1,500 rpm for 2 min and resuspended in 200 µl PBS containing 1% bovine serum albumin (BSA). After FcR blocking with a 1:200 dilution of anti-CD16/32 (BioLegend, San Diego, CA, USA) for 15 min, cells were stained for 30 min at 4°C with 25 µl of 1:150 dilution of the following antibodies: anti-CD11b-V450 (BD Biosciences, San Jose, CA, USA), anti-F4/80-PE (BioLegend), anti-Ly6G-PerCP-Cy (BD Biosciences), and anti-CD45-APC-Cy7 (BD Biosciences) for 30 min on ice in the dark. Cells were washed with PBS with 1% BSA, then fixed with 4% PFA. FACSVerse (BD) was used for flow cytometry analysis. After excluding dead cells and debris by gating with forward and side scatter, neutrophils (polymorphonuclear cells; PMNs) were detected as CD11b⁺, Ly6G⁺, and CD45⁺ components. Macrophages were detected as F4/80⁺, Ly6G^-, and CD45⁺ components. The absolute number of cells in each sample was counted.

The blood samples were centrifuged at 5,000 rpm for 5 min and the sera were collected for enzyme-linked immunosorbent assay (ELISA). Quantitative evaluation of cytokines (TNF-α, IL-6, and IL-1β) in serum was performed by following the ELISA kit protocol (Proteintech, Rosemont, IL, USA).

To assess the impact of TRPV1 and TRPV4 on the neutrophil phagocytic function, we performed bactericidal assays using murine PMNs ex vivo as previously reported (25). Bone marrow cells were flushed from the 6-week-old murine femora and tibia with RPMI 1640 (Thermo Fisher Scientific, Waltham, MA, USA) with 10% fetal bovine serum and 2 mM EDTA. By centrifugation for 30 min at 2,000 rpm, PMNs were isolated with a discontinuous gradient of Histopaque-1077 and -1119 (Sigma-Aldrich, St. Louis, MO, USA). The cells were then resuspended with Hank’s Balanced Salt Solution (Thermo Fisher Scientific) and adjusted to the desired concentration. We confirmed that more than 90% of the CD45⁺ cells isolated were Ly6G⁺ and CD11b⁺. The bacteria (10² CFU) were pre-opsonized with baby rabbit complement (Pel-Freez Biologicals, Rogers, AR, USA) for 30 min and applied to the PMNs in a bacterium: PMN ratio of 1:1,000. After 45 min of incubation at 37°C, the bacteria were incubated on the blood agar overnight and the number of colonies counted for quantification. The bacterial survival rate was calculated by dividing the number of bacteria in the experimental sample by those in the control that did not contain PMNs.

All analyses were performed by GraphPad Prism 8.4.3 (GraphPad Software, San Diego, CA, USA). Analysis of the survival experiment was performed by the Kaplan–Meier log-rank test. The Kruskal–Wallis test with Dunn’s multiple-comparison test was applied to compare the three groups. p < 0.05 was considered to indicate a significant difference.

To evaluate the role of TRPV1 and TRPV4 for pneumococcal carriage, mice (wild-type, TRPV1 KO, TRPV4 KO) were intranasally inoculated with two different doses (1 × 10⁵ and 1 × 10⁷ CFU/mouse) of S. pneumoniae 6A strain without anesthesia ( Figure 1 ). S. pneumoniae was successful in establishing nasal colonization at the same level in all mouse strains ( Figures 2A, C ). The density of colonization was dose-dependently increased among all mouse strains. At the same timing as nasal lavages, the blood cultures were also evaluated for detection of bacteremia. In the low-dose group, bacteremia was not detected in any mouse strains ( Figure 2B ). On the other hand, a significantly higher ratio of bacteremia was observed in TRPV4 KO mice (15 out of 31 mice, 48.4%) compared with wild-type (3 out of 20 mice, 15%) and with TRPV1 KO mice (4 out of 14 mice, 28.6%) by high-dose inoculation ( Table 1 ). The density of bacteremia in TRPV4 mice was significantly higher than that of wild type by Kruskal–Wallis test (p = 0.028) ( Figure 2D ).

Next, we evaluated the impact of influenza virus coinfection, one of the most exacerbating factors of pneumococcal diseases, on nasal colonization and bacteremia. Even when the mice had been infected with influenza A virus (IAV), the density of nasal colonization was almost equal among all mouse strains, regardless of the dose of inoculum ( Figures 3A, C ). Unlike the pneumococcal mono-infection model, bacteremia was observed, even with the low-dose inoculum. Both the density and the incidence of bacteremia were significantly higher in TRPV4 mice than in wild-type mice (p = 0.036) ( Figure 3B ). Although bacteremia was detected at the same frequency in all mouse strains when the dose of inoculum was increased ( Table 1 ), TRPV4 KO mice showed the highest density of bacteremia ( Figure 3D ). In this mouse model using a virulent pneumococcal strain, TRPV4 KO mice were more susceptible to development of bacteremia, although all mouse strains were colonized equally.

The local status of inflammation was analyzed with lavages of the upper respiratory tract. The number of PMNs in the nasal lavages on day 2 was counted by flow cytometry. The influx of PMNs was suppressed in TRPV4 KO mice compared to TRPV1 KO and wild-type mice ( Figures 4A, C ). Especially in the coinfection group, the number of PMNs was significantly lower in TRPV1 KO (p = 0.043) and TRPV4 KO (p = 0.0002) than that of wild type. Among TRPV4 KO mice, the number of PMNs was significantly lower in the coinfection model than in the mono-infection model (p = 0.018 by Mann–Whitney’s U test), which was the opposite result of an increase of PMNs in the coinfection model among wild-type mice.

We also counted the number of macrophages in the nasal lavages. There was a significant difference between TRPV1 KO and wild type in the mono-infection model ( Figure 4B ). In the coinfection model, there were no differences in the number of macrophages among the three mouse strains ( Figure 4D ).

To see the spread of the pneumococcus to various organs after nasal colonization, the mice were sacrificed on day 3 and the hearts, spleens, lungs, and brains were collected for culture ( Figure 1 ). In the pneumococcal mono-infection model, we could not find bacteria in any organs in any mouse strains (data not shown). Contrary to wild type, the bacteria were detected in most of the samples of TRPV1 KO and TRPV4 KO mice in the coinfection model ( Figure 5 ). Especially in the TRPV4 KO mice, the density of bacteria in all organs examined was significantly increased as compared with the wild type.

To investigate the route of pneumococcal infection in the multiple organs, the lung tissues were histologically analyzed. In the mono-infection model, no inflammation was observed in the lungs among all mouse strains ( Figures 6A – F ). In the coinfection model, the lungs of TRPV4 KO mice showed inflammatory infiltrates with neutrophils and macrophages in the interstitium and hemorrhage in the alveoli ( Figures 6K, L ). Furthermore, the lung tissues of TRPV1 KO mice demonstrated alveolar walls showing capillary congestion and mild to moderate inflammatory infiltrate including inflammatory cells ( Figures 6I, J ). On the other hand, no evidence of apparent inflammation was observed in the lung tissues of wild-type mice ( Figures 6G, H ). The severity of inflammation was quantified using a previously reported scoring system (24). There was a significantly higher inflammation in TRPV4 KO mice than wildtype when the mice were infected with both S. pneumoniae and IAV ( Figure 6M ). We also evaluated the lung tissues of the mice infected only with IAV and confirmed no apparent findings of viral pneumonia (data not shown). Histological findings suggest that TRPV1 and TRPV4 may be involved in regulating the pneumococcus migration to the lower respiratory tract and dissemination to multiple organs.

To evaluate the role of TRPV1 and TRPV4 on pneumococcal invasive diseases, we established a spontaneous pneumococcal lethal infection model after nasal colonization. The mice were given the bacteria intranasally on day 0 of the experiment with the dose of 1 × 10⁷/mouse. They were monitored every 12 h until they showed signs of lethal infection, such as decreased motor activity, shivering, and messy hair ( Figure 1 ). Similar time courses were observed, and around 70% of mice finally developed lethal infection among all groups in the mono-infection model (wild type; 65%, TRPV1 KO; 62%, TRPV4 KO; 75%, respectively) ( Figure 7A ).

When the mice were pre-infected with IAV, more than 70% of TRPV1 KO mice eventually developed lethal infection. TRPV4 KO mice showed the worst course in the coinfection model, and all mice developed lethal infection within 7 days after the pneumococcal challenge. Wild type showed significantly better survival (50%) than TRPV4 KO mice ( Figure 7B ). We also observed the mice infected with only IAV, and none of the mice in any of the three strains died during the observation period (data not shown).

Blood samples were collected from the colonized mice on day 2 of the experiment. The sera were separated by centrifugation, and the levels of cytokines—tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β)—were measured by ELISA. In the mono-infection model, the levels of TNF-α were significantly higher in TRPV4 KO mice than those of wild type and TRPV1 KO mice (p < 0.05 and p < 0.001, respectively) ( Figure 8A ). Similar levels of IL-6 and IL-1β were observed among wild-type and TRPV4 KO mice, whereas the production of IL-6 and IL-1β was low in TRPV1 KO mice ( Figures 8B, C ).

When coinfected with IAV, the production of TNF-α was greatly suppressed in TRPV4 KO mice compared with the mono-infection model ( Figure 8D ). The levels of IL-6 and IL-1β were lower in TRPV1 KO and TRPV4 KO mice than in wild-type mice, although without statistical differences ( Figures 8E, F ).

To evaluate the mechanism of protection against pneumococcal infection, we focused on the phagocytic bactericidal ability of neutrophils, which is one of the most effective systems for eliminating the pneumococcus. Neutrophils were isolated from each mouse strain and incubated with pre-opsonized bacteria with baby rabbit complement. The survival rate of the bacteria was compared between three groups. The experiment was repeated three times for each mouse strain, and the pooled data are displayed in Figure 9 . Neutrophils derived from TRPV4 KO mice showed significantly lower bactericidal ability than those derived from wild type (p < 0.0001). TRPV1 KO mice showed a similar tendency, but no significant difference was observed.