Analysis of Immune Responses in Acinetobacter baumannii-Infected Klotho Knockout Mice: A Mouse Model of Acinetobacter baumannii Infection in Aged Hosts

Acinetobacter baumannii is an important opportunistic pathogen that primarily afflicts elderly people. To clarify the pathogenicity of A. baumannii in the elderly, we investigated immune responses to A. baumannii ATCC 19606 infection in klotho knockout (KO) mice, the mouse model of aging. Following intravenous inoculation, the mice seldom displayed severe symptoms. However, the survival rate was 56% at 7 days post-infection. Bacteria were detected in the lungs of klotho KO mice but not klotho wildtype (WT) mice at 7 days post-infection. Neutrophils, eosinophils, interstitial macrophages, and monocyte/dendritic cell subset in the lungs of klotho KO mice were transiently induced after infection with A. baumannii. The number of alveolar macrophages in klotho KO mice was lower than that in klotho WT mice, except for 1 day post-infection. CD11b expression on neutrophils and alveolar macrophages in the lungs of klotho KO mice was seldom upregulated by the infection. These results suggested that immune functions eliminating bacteria in the lungs of klotho KO mice were insufficient. CD11blow conventional DC cells hardly increased in klotho KO mice infected with A. baumannii. Additionally, the production of interleukin (IL)-10 in the sera of klotho KO mice was significantly higher than that in klotho WT mice, whereas that production of interferon-gamma was not detected in the sera of klotho KO mice. These results suggested that acquired immune responses were hardly induced in klotho KO mice. IL-1β, CXCL1, CXCL2, and CCL2 expression was significantly higher in the lungs of klotho KO mice infected with A. baumannii than in those of klotho WT mice at 1 day post-infection. These results suggested that pulmonary inflammation was elicited in klotho KO mice during early infection. The expression levels of proinflammatory cytokines significantly correlated with TLR9 expression in the lungs of klotho KO mice. The collective results demonstrate an A. baumannii infection state in aged hosts and suggest that pulmonary inflammation and bacterial burden should be noted in aged hosts even in the absence of severe symptoms of A. baumannii infection.


INTRODUCTION
Acinetobacter baumannii is an important opportunistic pathogen that is associated with nosocomial infections that include bacteremia, pneumonia, meningitis, urinary tract infections, and wound infections (1,2). The recent increase in multidrugresistant A. baumannii (MDRAB) outbreaks worldwide has become a concern (3)(4)(5). Additionally, A. baumannii is among the six nosocomial pathogens-Enterococcus faecium, Staphylococcus aureus, Klebsiella pneumoniae, A. baumannii, Pseudomonas aeruginosa, and Enterobacter spp.-that comprise the ESKAPE group that have acquired multidrug resistance and virulence (6,7). A. baumannii is also an emerging pathogen among elderly people in community hospitals and nursing homes (8). Given the accelerating pace of aging of the global population, A. baumannii is an important global human pathogen.
Although A. baumannii is regarded as a low-virulence pathogen (9), it possesses several mechanisms of pathogenicity, including biofilm formation, adherence, and invasion of lung epithelial cells (10)(11)(12)(13)(14), host cell death (15)(16)(17), and iron acquisition (18). The pathogenicity of A. baumannii depends on various virulence factors, especially the outer membrane proteins (Omps) (19). Additionally, we have reported that MDRAB clinical isolates are resistant to toxicity caused by reactive oxygen species (ROS) in macrophages (20). Taken together, A. baumannii is an emerging, highly pathogenic bacterium with characteristics that vary with environmental stress conditions, such as multiple antimicrobial agents and host immune responses. However, only few studies have explored the reasons for the pathogenicity of A. baumannii among elderly people.
Some studies on host immune responses to A. baumannii infection using mouse models have been performed. Neutrophils and alveolar macrophages (AM) play important roles in host resistance to respiratory infection of A. baumannii (21)(22)(23). In addition, Toll-like receptor (TLR) 2 and 4 play key roles in eliciting innate immune responses against A. baumannii infection (24,25). TLR9 contributes to the expression of proinflammatory cytokines and chemokines (26). These results suggest that innate immune responses play important roles in host resistance to early A. baumannii infection. However, these studies have focused on the immune responses of young hosts against A. baumannii infection. In contrast, aged mice reportedly showed increased susceptibility to A. baumannii infection, increased inflammatory cell infiltration, increased proinflammatory cytokine levels at 24 h post-infection, as well as impaired efficacies of antibiotics and a vaccine (27). These results suggest that immune responses are altered in aged mice compared with the responses in young mice.
Klotho protein is senescence suppressor and its expression naturally reduces in the elderly and in individuals with choric kidney disease (28). Klotho knockout (KO) mice show several aging phenotypes that are similar to human aging (29). These phenotypes include arteriosclerosis, osteoporosis, age-related skin changes, and ectopic calcifications, together with short lifespan and infertility (29). A klotho-septic mouse model that displayed hypercytokinemia with impaired bacterial clearance and increased apoptosis of lymphocytes may be related to poor survival in klotho KO mice treated with cecal ligation and puncture (30). Additionally, a previous study reported that klotho deficiency aggravates sepsis-related multiple organ dysfunction (31). Thus, klotho KO mice are expected to be useful as an animal model for the clarification of pathogenesis of bacterial infections in aged hosts. Generally, aging affects multiple immune systems (32)(33)(34). However, although previous studies have reported acute immune responses to A. baumannii infection in young mice (35)(36)(37), only a few studies have investigated the immune responses of aged hosts to A. baumannii infection. Additionally, although immune responses are altered in aged mice compared with those in young mice at 24 h post-infection (27), the immune responses that occur for a few days after A. baumannii infection in aged hosts remain unknown.
In this study, we established an A. baumannii infection klotho KO mouse model via intravenous inoculation and evaluated its immune responses upon infection with A. baumannii. Since TLRs play an important role in eliciting immune responses to A. baumannii infection (24)(25)(26), we also analyzed the correlation between the expression levels of proinflammatory cytokines and those of TLRs in the lungs of the KO mice.

Bacterial Strains and Growth Condition
ATCC 19606 strains of A. baumannii were used in this study. Bacteria were cultured on Luria-Bertani (LB) agar plates (Becton, Dickinson and Company, Sparks, MD, USA) for 16 h at 37°C. The bacteria were washed twice with PBS and resuspended at a concentration of 5 × 10 8 colony forming units (CFU)/ml, with the concentration adjusted via optical density measurements at 595 nm. The bacterial suspensions obtained were used for animal experiments. All procedures were carried out in accordance with the relevant guidelines and regulations.

Animal Experiments
All animal experiments were performed in accordance with the Institutional Animal Care and Use Committee of Teikyo University Animal Ethics Committee (approval #17-009). Fiveweek-old male klotho KO mice and littermate klotho wildtype (WT) mice were purchased from CLEA Japan, Inc. (Tokyo, Japan). Bacteria (10 8 CFU/200 ml) were inoculated intravenously into 6-week-old klotho mice. PBS were inoculated intravenously into 6-week-old klotho mice as uninfected mice. To analyze the number of bacteria in the lungs and spleen of klotho mice, 100 ml of the homogenates in PBS was plated on LB agar plates. The bacterial count (CFU) was confirmed by the growth of serial dilutions of the bacterial suspension on LB agar after 24 h of incubation at 37°C. To analyze the number of bacteria in the blood of klotho mice, 100 ml of blood was plated on LB agar plates. The bacterial count was confirmed using the same method. mice were gently killed by cervical dislocation under sedation, and the lungs and spleens were surgically harvested. The lung tissue was cut into small pieces with scissors, and processed in digestion buffer (200 mg/ml of Collagenase Type 1 CLS1 (Worthington Biochemical, Lakewood, NJ) and 2 mM of CaCl 2 in PBS). The homogenized lungs and spleen were prepared by gently pressing the organs through a 40-mm nylon mesh into a single-cell suspension. The cells were washed and red blood cells were lysed with PharmLyse Lysis Buffer (BD Biosciences, San Jose, CA). The harvested blood was treated with PharmLyse Lysis Buffer (BD) and the cells were washed with PBS. Isolated cells were counted and then stained with monoclonal antibodies (mAbs) for the analysis of immune cells. Briefly, isolated cells from the lungs were stained with various combinations of mAbs. The mAbs included were fluorescein isothiocyanate (FITC)labeled anti-F4/80, phycoerythrin (PE)-labeled anti-CD11b, PE-Cy-7-labeled anti-CD45, allophycocyanin (APC)-labeled anti-CD11c, and APC-Cy-7-labeled Ly-6G mAbs (Biolegend, San Diego, CA). The white blood cells from the spleen and blood were stained with various combinations of mAbs, including FITC-labeled anti-F4/80, PE-labeled anti-CD11b, PE-Cy-7-labeled anti-Ly-6C, APC-labeled anti-CD11c, and APC-Cy-7-labeled Ly-6G mAbs. To analyze the proportion of immune cells in the lungs, spleen, and blood of klotho mice, immune cells were stained with specific antibodies against surface markers for 30 min at 4°C. The cells were washed twice with stain buffer (BD), and subsequently fixed with BD Cytofix ™ Fixation Buffer (BD). The stained cells were analyzed using a FACSCant II flow cytometer equipped with FACS Diva software (BD). All flow cytometry data were analyzed using FlowJo software (BD).

RNA Extraction and Quantitative Real-Time Polymerase Chain Reaction (qPCR)
To analyze the expression of proinflammatory cytokines, total RNA was extracted from the lungs of uninfected and A. baumannii-infected klotho mice using the RNeasy Plus Mini kit (QIAGEN, Tokyo, Japan). Harvested RNA samples were quantified using the NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA). Total RNA was reversetranscribed to cDNA using PrimeScript ™ 1 st strand cDNA Synthesis Kit (TaKaRa Bio, Shiga, Japan). To analyze the mRNA levels of all genes, cDNA was amplified using the PowerUP SYBR Green Master Mix (Thermo Fisher Scientific) with consensus primers for detecting mouse glyceraldehyde 3phosphate dehydrogenase (GAPDH), mouse TNF-a, mouse IL-1b, mouse IL-6, mouse IL-10, mouse CXCL1, mouse CXCL2, mouse CCL2, mouse CCL7, mouse TLR2, mouse TLR4, and mouse TLR9. The primer sequences are listed in Table 1.
Mouse GAPDH was used as an internal control to quantify mouse TNF-a, mouse IL-1b, mouse IL-6, mouse IL-10, mouse CXCL1, mouse CXCL2, mouse CCL2, mouse CCL7, mouse TLR2, mouse TLR4, and mouse TLR9. Real-time PCR was performed using 40 cycles of denaturation at 95°C for 15 s, annealing at 62°C for 15 s, and extension at 72°C for 1 min. The amplified PCR products were quantitatively monitored using a StepOne Real-Time PCR System (Applied Biosystems, Foster City, CA). Fold changes in the expression level of each gene were calculated by the 2 −DDCt method using the mouse GAPDH gene as an internal control. The relative expression of each gene was evaluated relative to the control sample (uninfected mouse), which was assigned an arbitrary unit value of 1.

Statistical Analyses
Survival curves with log rank test were obtained using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics (42). Quantitative results were compiled from more than two independent experiments. The results are presented as the mean ± standard error of the mean (SEM). Comparisons of numerical data were performed using the Student's t-test. Pearson correlation analysis was used to compare the mRNA levels of proinflammatory cytokines and TLR2, 4, and 9. In all analyses, a two-tailed probability of < 5% (i.e., *P < 0.05) was considered statistically significant.  Figure  1A).  symptoms. Additionally, although klotho WT mice did not die of A. baumannii infection, klotho KO mice were more susceptible, as evidence by the 56% survival rate at 7 days post-infection ( Figure 1C). These results suggested that klotho KO mice rarely displayed severe symptoms of A. baumannii infection prior to their death. To clarify whether bacteria were eliminated from klotho KO mice, the bacteria in the lungs, spleen, and blood of klotho WT and KO mice were enumerated after infection with A. baumannii. Bacteria were not detected in the lungs, spleen, and blood of klotho WT mice at 7 days post-infection ( Figures 1D-F; WT). In contrast, bacteria were detected in the lungs of three of four klotho KO mice at 7 days post-infection ( Figure 1D;  klotho WT mice significantly increased at 1 day post-infection ( Figure 2A). The number of eosinophils in the lungs of klotho KO mice significantly increased at 1 day post-infection and subsequently decreased until 7 days post-infection, whereas klotho WT mice showed no induction of eosinophils after infection ( Figure 2B). The number of AM in the lungs of klotho KO mice was slightly increased at 1 day post-infection and subsequently decreased to the pre-infection level at 3 days post-infection, whereas the cell number in klotho WT mice slightly decreased after infection ( Figure 2C). Additionally, the number of AM in klotho KO mice was lower than that in klotho WT mice, except for 1 day post-infection ( Figure 2C). The number of IMs increased in the lungs of klotho KO mice at 1 day post-infection and subsequently deceased at 3 days postinfection. ( Figure 2D). The number of CD11b +/hi CD11c −/low monocytes plus CD11b hi CD11c low Mc/DC subset in the lungs of klotho KO mice increased at 1 day post-infection and subsequently decreased to the pre-infection level by 7 days post-infection, whereas that of klotho WT mice gradually increased until 3 days post-infection and was decreased at 7 days post-infection ( Figure 2E). The number of CD11b low conventional DC in the lungs of klotho KO mice hardly increased after the infection, whereas that of klotho WT mice increased at 3 days post-infection and subsequently decreased at 7 days post-infection ( Figure 2F). These results suggested that innate immune cells in the lungs of klotho KO mice were transiently induced after infection of A. baumannii. However, this transient change did not eliminate A. baumannii in the lungs ( Figure 1D).

A. baumannii
Previous studies have reported that neutrophils are activated in inflammatory conditions as well as in response to lipopolysaccharide (LPS) stimulation, with the upregulation of markers of the integrin family, such as CD11b, as well as other activation markers (46,47). Although neutrophils were induced in klotho KO mice after infection with A. baumannii, the burden of the lung bacteria was not completely eliminated in the mice at 7 days post-infection. Therefore, we evaluated the expression level of CD11b on neutrophils to determine whether neutrophils were activated in the lungs of klotho KO mice after infection with A. baumannii.
Representative results from uninfected and A. baumanniiinfected klotho mice are shown in Figure 3A and the complete results are summarized in Figure 3B. The expression level of CD11b on neutrophils in the lungs of klotho WT mice was significantly increased after infection with A. baumannii ( Figures 3A, B; WT). In contrast, the expression level of CD11b on neutrophils in the lungs of klotho KO mice was slightly increased at only 1 day post-infection ( Figures 3A, B; KO). Moreover, there was a significant difference between the expression level of CD11b on neutrophils in the lungs of klotho KO mice and WT mice at 3 days post-infection ( Figure 3B). In addition, the expression level of Ly-6G as a maturation marker of neutrophils was analyzed in klotho WT and KO mice after infection with A. baumannii. The expression level of Ly-6G on neutrophils in the lungs of klotho WT mice gradually increased after the infection, whereas that of klotho KO mice hardly increased after infection with A. baumannii ( Figures 3C, D).
There was a significant difference between the expression level of Ly-6G on neutrophils in the lungs of klotho WT mice and KO mice at 3 days post-infection. These results suggest that neutrophils in the lungs of klotho KO mice impair their activation and functional maturation. We further analyzed the expression level of CD11b on AM, because the expression is upregulated by host-derived inflammatory stimuli (48,49). Representative results from uninfected and A. baumanniiinfected klotho mice are shown in Figure 3E and the collective results are summarized in Figure 3F.  Figure 4A). Additionally, the level of TNFa mRNA in the lungs at 3 days post-infection was significantly lower in klotho KO mice than in klotho WT mice ( Figure 4A). The level of IL-1b mRNA in the lungs of klotho WT mice significantly increased at 1 day post-infection and was maintained at 3 days post-infection, whereas those in the lungs of klotho KO mice strongly and significantly increased at 1 day post-infection and subsequently decreased at 3 days postinfection ( Figure 4B). The level of IL-1b mRNA was significantly higher in the lungs of klotho KO mice than in those of klotho WT mice at 1 day post-infection ( Figure 4B). The level of IL-6 mRNA in the lungs of klotho WT mice slightly increased at 1 day post-infection and subsequently deceased at 3 days post-infection ( Figure 4C). The level of IL-6 mRNA in klotho KO mice was not increased at 1 day post-infection and was decreased at 3 days post-infection ( Figure 4C). The level of IL-10 mRNA before infection was significantly higher in the lungs of klotho KO mice than in those of klotho WT mice ( Figure 4D). The level of IL-10 mRNA in the lungs of klotho WT mice significantly increased at 1 day post-infection and decreased at 3 days post-infection ( Figure 4D). The level of IL-10 mRNA in the lungs of klotho KO mice increased at 1 day post-infection and subsequently decreased significantly ( Figure 4D). The level of CXCL1 mRNA in the lungs of klotho WT mice gradually increased after infection ( Figure 4E). The level of CXCL1 mRNA in the lungs before infection was higher in klotho KO mice than in klotho WT mice ( Figure 4E). Additionally, the level of CXCL1 mRNA in the lungs at 1 day post-infection was significantly higher in klotho KO mice than in klotho WT mice ( Figure 4E). The level of CXCL2 mRNA in the lungs of klotho KO mice  increased at 1 day post-infection and subsequently decreased at 3 days post-infection ( Figure 4F). The level of CXCL2 mRNA was significantly higher in the lungs of klotho KO mice than in those of klotho WT mice at 1 day post-infection ( Figure 4F). The level of CCL2 mRNA in the lungs of klotho WT mice gradually increased after infection, whereas those of klotho KO mice significantly increased at 1 day post-infection and subsequently decreased at 3 days post-infection ( Figure 4G). The level of CCL2 mRNA in the lungs at 1 day post-infection was significantly higher in klotho KO mice than in klotho WT mice ( Figure 4G). The level of CCL7 mRNA increased in the lungs of klotho WT and KO mice at 1 day post-infection and subsequently decreased at 3 days post-infection ( Figure 4H). The expression levels of CCL7 mRNA were comparable between klotho WT and KO mice after infection ( Figure 4H). The collective results suggested that proinflammatory responses were strongly and transiently induced in the lungs of klotho KO mice after infection with A. baumannii. Bar graph data represent the mean ± SEM and were compiled from four independent experiments (n = 12 for WT and n = 12 for KO mice). Asterisks indicate statistically significant differences (**P < 0.01; *P < 0.05, before infection vs. after infection, Student's t-test; † † P < 0.01; † P < 0.05, klotho WT mice vs. klotho KO mice, Student's t-test).

Sato et al. Immune Responses to A. baumannii
Frontiers in Immunology | www.frontiersin.org November 2020 | Volume 11 | Article 601614 analyzed the expression levels of TLR2, 4, and 9 in the lungs of klotho WT and KO mice after infection with A. baumannii. The level of TLR2 mRNA was significantly lower in the lungs of uninfected KO mice than in those of uninfected klotho WT mice ( Figure 4I). Following infection, the level of TLR2 mRNA in the lungs of klotho WT mice was stable but was significantly increased in KO mice at 1 and 3 days post-infection. The level of TLR2 mRNA was significantly lower in the lungs of klotho KO mice than in those of klotho WT mice at 3 days post-infection. In addition, the level of TLR9 mRNA was lower in the lungs of uninfected KO mice than in those of klotho WT mice ( Figure  4K). Following infection, the levels of TLR9 mRNA in the lungs of klotho WT and KO mice increased at 1 day post-infection and subsequently decreased at 3 days post-infection. However, the level of TLR9 mRNA in the lungs of klotho KO mice was significantly decreased compared with that of klotho WT mice at 3 days post-infection. In contrast, the levels of TLR4 mRNA were comparable between klotho WT and KO mice, and were not altered in their lungs after the infection ( Figure 4J). We further analyzed the correlation between the expression levels of proinflammatory cytokines and those of TLRs in the lungs of klotho WT and KO mice after infection with A. baumannii.
Pearson correlation analysis revealed that the mRNA level of CXCL2 was significantly correlated with that of TLR2 in the lungs of klotho WT mice ( Figure 5F). In contrast, mRNA levels of TNF-a and IL-1b were significantly correlated with those of TLR2 in the lungs of klotho KO mice ( Figures 5A, B).
Additionally, mRNA levels of IL-6, IL-10, CXCL1, CCL2, and CCL7 were not correlated with those of TLR2 in the lungs of klotho WT and KO mice (Figures 5C-E, G, H). Moreover, mRNA levels of CCL2 and CCL7 were significantly correlated with those of TLR9 in the lungs of klotho WT mice ( Figures 6G,  H). In contrast, the mRNA levels of TNF-a, IL-1b, IL-10 CXCL1, CXCL2, CCL2, and CCL7 were strongly and significantly correlated with those of TLR9 in the lungs of klotho KO mice ( Figures 6A, B, D-H). Additionally, mRNA level of IL-6 was not correlated with that of TLR9 in the lungs of klotho WT and KO mice ( Figure 6C). These results indicated the important roles of TLR9 in eliciting proinflammatory responses in the lungs of aged hosts against A. baumannii infection.

Different Immune Cell Induction Pattern in the Blood of Klotho KO Mice After Infection With A. baumannii
We analyzed the number of immune cells in the blood of klotho WT and KO mice after infection with A. baumannii. Representative results from uninfected and A. baumanniiinfected klotho mice at 1 day post-infection are shown in Figures S3 (WT) and S4 (KO). The results from the two groups are summarized in Figures 7A-D. Flow cytometry revealed that the number of neutrophils in the blood before infection was lower in klotho KO mice than in klotho WT mice ( Figure 7A). The number of neutrophils in the blood of klotho WT mice significantly decreased at 1 day-post infection and subsequently increased gradually ( Figure 7A). In contrast, the number of neutrophils in the blood of klotho KO mice increased at 3 days post-infection and subsequently decreased at 7 days post-infection ( Figure 7A). The number of eosinophils significantly decreased in the blood of klotho WT and KO mice at 1 day-post infection and subsequently increased ( Figure 7B). The number of monocytes in the blood of klotho KO mice significantly increased at 1 day post-infection and subsequently decreased at 3 days post-infection ( Figure 7C). However, the number of monocytes in the blood before and after infection was lower in klotho KO mice than in klotho WT mice ( Figure 7C). The number of macrophages in the blood of klotho KO mice significantly increased at 1 day post-infection and subsequently decreased gradually, whereas those of klotho WT mice increased at 3 day post-infection and subsequently decreased at 7 days post-infection ( Figure 7D).

Klotho KO Mice Infected With A. baumannii Display Different Cytokine Production Pattern in Their Sera
We analyzed the levels of proinflammatory cytokines in the sera of klotho WT and KO mice before and after infection with A. baumannii. TNF-a, IL-6, CXCL1, CCL2, and CCL7 levels significantly increased in the sera of klotho WT and KO mice at 1 day post-infection, while the level of IL-12 was not ( Figures  7E, F, I-K). Of note, although the serum level of IL-6, CXCL1, and CCL7 in klotho WT mice decreased at 3 days post-infection, those in klotho KO mice was maintained until 3 days postinfection ( Figures 7F, I, K). The serum level of IL-10 increased in klotho WT and KO mice after infection and was significantly higher in klotho KO mice than in klotho WT mice at 1 day postinfection ( Figure 7G). The serum level of IFN-g significantly increased in klotho WT mice at 1 day post-infection, whereas that did not increase in infected klotho KO mice ( Figure 7H). These results suggested that proinflammatory responses were prolonged in the periphery of klotho KO mice infected with A. baumannii.

DISCUSSION
A. baumannii has recently emerged as a major nosocomial pathogen (1,2). Although we have previously reported the virulence characteristics of A. baumannii clinical isolates in human epithelial cells and macrophages in vitro (13,14,20), the pathogenicity of A. baumannii infection in elderly people has remained elusive. In the current study, we established an aged mouse model of A. baumannii infection using klotho KO mice and evaluated the immune responses in the mice before and after infection. Previous study reported an aged pneumonia mouse model by non-invasive intratracheal inoculation with A. baumannii LAC-4 strain (27). Higher mortality was observed in aged mice along with increased bacterial burdens and more severe lung injury. In addition, increased inflammatory cell infiltration and enhanced pro-inflammatory cytokines at 24 h post-infection were detected in aged mice. A. baumannii LAC-4 strain is much more virulent than A. baumannii ATCC 19606 strain (36). As A. baumannii   (52). These results suggest that acute inflammation is elicited in the lungs of aged hosts early during A. baumannii infection. Inflammatory monocytes are recruited into tissues and kill pathogens via phagocytosis, ROS activity, production of nitric oxide (NO), and the action of inflammatory cytokines (53,54). Although we analyzed ROS production in monocytes as well as neutrophils in klotho KO mice infected with A. baumannii, it was not clear whether those cells induced ROS in response to the infection. Further studies are required to clarify the production of ROS in those cells. On the other hands, inflammatory monocytes upregulate levels of CD11c and MHC class II, which can promote T cell proliferation (53). In the present study, inflammatory monocytes in the lungs, spleen ( Figures  S5, S6, and S7), and blood (Figures 7, S3, and S4) of klotho KO mice infected with A. baumannii were transiently induced. However, the number of inflammatory monocytes in those tissues were lower in klotho KO mice than in klotho WT mice. These results suggest that A. baumannii infection is complicated by the low induction of inflammatory monocytes in aged hosts. Additionally, klotho KO mice infected with A. baumannii showed lower induction of DC subsets than infected klotho WT mice (Figures 2 and S7). These results suggest that ageassociated changes in the proportion of monocyte and DC subsets affect adaptive immune function, thereby reducing the efficacies of antibiotics and vaccines in aged hosts (27).
Although the present study suggests the disruption in the function and/or differentiation of immune cells in klotho KO mice, it remains unclear how klotho KO mice died from infection with A. baumannii. A previous study reported that klotho deficiency aggravates sepsis-related multiple organ dysfunction (31). Thus, one possible reason for eventually death might be incomplete bacteria clearance and worsen multiple organ dysfunction. Hearps et al. reported that monocytes from older individuals exhibit impaired phagocytosis but contain shortened telomeres, suggesting a dysregulation of monocytes function in the aged hosts (55). Moreover, it is suggested that telomere shortening in peripheral monocytes reflects telomere shortening in haematopoietic bone marrow precursor cells (55). Considering the above findings, impaired activation of immune cells in klotho KO mice may be ascribed to telomere shortening in haematopoietic bone marrow precursor cells and causes incomplete bacteria clearance in the lungs, and then led to death with sepsis-related multiple organ dysfunction.
IL-1b is a potent proinflammatory cytokine. Its overexpression can lead to tissue damage (56). For instance, decreased IL-1b in inflammatory responses leads to a better outcome in acute lung infections with Pseudomonas aeruginosa (57). Pulmonary inflammation in mice after infection with A. baumannii has been associated with IL-1b production and the severity of lung pathology (58). In addition, CC and CXC chemokines attract neutrophils and mononuclear cells to sites of inflammation (59)(60)(61)(62). Mediators of inflammation, including IL-1 and TNF-a, or bacterial products such as LPS elicit the production of those chemokines (59). Presently, the induction level of IL-1b, CXCL1 (KC), CXCL2 (MIP-2), and CCL2 (MCP-1) mRNA in the lungs after infection was significantly higher in klotho KO mice than in klotho WT mice. This result suggests that klotho KO mice infected with A. baumannii showed severe lung inflammation compared with klotho WT mice, which sometimes led to death. A previous study reported that TLR2 KO mice displayed increased lung concentrations of MCP-1 and MIP-2 after infection with A. baumannii (24). The authors discussed the possibility that TLR2 mediates anti-inflammatory pathways that downregulate MCP-1 production. We also demonstrated that the expression level of TLR2 mRNA was lower in the lungs of uninfected klotho KO mice than in those of uninfected klotho WT mice. However, the expression level in the lungs of klotho KO mice was upregulated at 1 day post infection of A. baumannii. These results suggest the possibility that initial infection of A. baumannii in aged hosts causes severe inflammation. Moreover, the expression level of TLR9 mRNA in the lungs of klotho KO mice was lower than that of klotho WT mice. TLR9 KO mice challenged intranasally with A. baumannii reportedly displayed a significantly increased bacterial burden and neutrophil recruitment in the lungs, with reduced levels of proinflammatory cytokines, such as TNF-a, IFN-g, and MCP-1, in their sera during systemic A. baumannii infection (26). These phenotypes are similar to those of klotho KO mice infected with A. baumannii in the present study. Additionally, we demonstrated that the expression of proinflammatory cytokines was strongly and significantly correlated with that of TLR9 mRNA in the lungs of klotho KO mice. The above findings suggest that TLR9-mediated immune responses may play an important role in the lung inflammation of elderly people with A. baumannii infection.
In this study, lung bacterial burden remained in klotho KO mice at 7 days post-infection in spite of the infection established by the intravenous inoculation of A. baumannii. Previous studies have demonstrated that the virulence factors of A. baumannii, such as OmpA and Bap, mediate bacterial adherence to bronchial and lung epithelial cells (11,12). Additionally, we have reported that the levels of ompA expression in MDRAB clinical isolates correlate with the adherence of the bacteria to lung epithelial cells (13). We also reported a novel bacterial transport mechanism, where A. baumannii exploits human neutrophils by adhering to the cells (63). These results support the novel idea concerning A. baumannii infection in elderly people, in which the bacteria infect aged hosts via the peripheral blood, respiratory system, and other routes, followed by the movement with neutrophils to the lung tissue and subsequent colonization of the lungs. Additionally, the production of proinflammatory cytokines in the sera after infection with A. baumannii were prolonged in klotho KO mice compared with klotho WT mice. However, although klotho WT mice showed decreased weight after infection with A. baumannii, klotho KO mice showed only a slight weight change after the infection. Additionally, most of klotho KO mice infected with A. baumannii did not display any clinical symptoms. These results suggest that weak inflammation lasting for prolonged periods are observed in aged hosts.
A previous study reported that A. baumannii infection inhibits airway eosinophilia and lung pathology in a mouse model of allergic asthma (64). Our results indicate that eosinophils are decreased in the lungs of klotho WT mice after infection with A. baumannii, implying the inhibition of airway eosinophilia in young hosts with allergic asthma. However, the significant increase in eosinophils in the lungs of klotho KO mice after infection with A. baumannii indicates further induction of airway eosinophilia in elderly people with allergic asthma. Thus, the pulmonary inflammation in A. baumannii infection in elderly people may be a different clinical condition compared with that in young people.
In summary, our results suggest that A. baumannii infection in aged hosts involves accumulation of the bacteria in the lungs, which elicits pulmonary inflammation during early infection and can eventually lead to death. The results also suggest that pulmonary inflammation correlates with altered TLR9 expression in klotho KO mice. This study provides insights into the pathogenicity of A. baumannii infection, particularly in elderly people. Further studies are required to understand the mechanism of A. baumannii infection in elderly people.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The animal study was reviewed and approved by Institutional Animal Care and Use Committee of Teikyo University Animal Ethics Committee (no. 1705236A1b).