Here, we present a protocol for establishing a mouse sepsis immunosuppression model via intranasal instillation of Klebsiella pneumoniae and validated the model through various indicators.
Method Article
June 12th, 2026
Here, we present a protocol for establishing a mouse sepsis immunosuppression model via intranasal instillation of Klebsiella pneumoniae and validated the model through various indicators.
Sepsis is a life-threatening organ failure caused by the host's inappropriate response to infection. Pneumonia is the primary cause of sepsis. Animal models are important tools for studying the pathogenesis of sepsis and evaluating new treatment strategies. Previous animal models of sepsis lacked natural infection pathways, making it difficult to accurately reflect the clinical infection process.
This study introduced a mouse sepsis immunosuppression model established by intranasal instillation of Klebsiella pneumoniae (KP) to simulate the natural respiratory infection pathway. After modeling with different concentrations of KP, the inflammation of the model was evaluated at 12 h for cytokine interleukin-6 (IL-6) levels, and organ damage was evaluated at 24 h for alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatine kinase (CK), and blood urea nitrogen (BUN) levels. The survival rate of the model was analyzed after 7 days. The percentage of lymphocytes and white blood cells, the proportion of major histocompatibility complex class II (MHC-II) expression in monocytes, and the mean fluorescence intensity (MFI) of IL-10 and TGF-β1 in CD4+ T cells were observed to evaluate the immune suppression in mice. After 7 days of modeling, a secondary infection was performed by intranasal instillation of Pseudomonas aeruginosa (PA). The bacterial load in bronchoalveolar lavage fluid (BALF) 24 h after secondary infection and the mortality rate within 14 days were assessed to further evaluate the model's immune suppression.
Nasal instillation of high-concentration Klebsiella pneumoniae can effectively induce an immune suppression. Its purpose is to simulate sepsis induced by a natural respiratory infection pathway and to provide a standard model for sepsis research in animals.
Sepsis is a life-threatening organ failure resulting from the host's inappropriate response to infection1,2,3. It is one of the leading causes of death in intensive care units (ICUs). Globally, there are 48.9 million sepsis cases annually, with 11 million associated deaths4. Despite recent advancements in anti-infective therapy and organ support technologies, the pathogenesis of sepsis remains incompletely understood, and specific treatment options are lacking. As such, sepsis has become a key focus and challenge in clinical medical research4.
Animal models play an irreplaceable role in basic and translational medical research, serving as essential tools for investigating sepsis pathogenesis and evaluating novel therapeutic strategies. By simulating the pathophysiological processes of human sepsis, these models not only facilitate exploration of key mechanisms such as immune disorders and organ dysfunction but also provide an experimental platform for drug screening and the development of intervention strategies. Among various animal models, mice are the most commonly used in sepsis research due to their well-defined genetic background, extensive research on their immune system, and ease of manipulation.
Currently, several sepsis animal models based on different paradigms are available, including endotoxin models, bacterial infusion models, cecal ligation and puncture (CLP) models, and colon ascendens stent peritonitis (CASP) models5. However, these models lack natural infection routes, making it challenging to truly reflect the clinical infection process. Studies have reported that pneumonia is the primary cause of sepsis in the United States6. Therefore, establishing a mouse sepsis model that mimics natural infection routes, exhibits immunosuppressive features, and is easy to operate with high reproducibility is of great significance for in-depth research into the immune mechanisms of sepsis and for evaluating potential intervention strategies.
Klebsiella pneumoniae (KP) accounts for 3% to 8% of all hospital-acquired bacterial infections7,8,9. This bacterium can enter the body through various portals, including the respiratory, gastrointestinal, and urinary tracts, as well as skin, and infections caused by KP have a poor prognosis, especially in immunocompromised patients7,10. Given that pneumonia is the leading cause of sepsis and KP is a common respiratory pathogen associated with poor outcomes, KP is an ideal candidate for establishing a clinically relevant sepsis model.
Although a variety of animal models of KP infection, including the intranasal method, have been established previously11, only a few studies have focused on optimizing the model to induce sepsis-associated immunosuppression or defining the optimal KP concentration for model establishment. This gap limits the utility of existing KP models for studying sepsis-induced immune dysfunction.
To address this need, we established a sepsis-induced immunosuppression model in mice via intranasal instillation of KP. Compared to existing models, this model simulates the natural pathways of respiratory infections. The model was further validated to provide a method for subsequent research.
All experimental procedures described in this article were approved by the Laboratory Animal Welfare and Ethics Committee of the Third Military Medical University (SCXK(PLA)20120031).
1. Experimental animals
2. Bacterial preparation
3. Intranasal instillation infection (Figure 1)
4. Model evaluation
NOTE: Four groups of mouse models were established via the aforementioned intranasal instillation method, namely the control group (group A, instilled with PBS), the 1 × 104 KP concentration group (group B), the 1 × 106 KP concentration group (group C), and the 1 × 108 KP concentration group (group D). The model was evaluated using the following methods:
At 12 h after intranasal instillation, the serum IL-6 level in the 1 × 108 KP group was significantly increased (p < 0.0001, n = 5) (Figure 2A). At 24 h after intranasal instillation, the serum levels of ALT (p < 0.01, n = 5) (Figure 2B), AST (p < 0.01, n = 5) (Figure 2C), CK (p < 0.01, n = 5) (Figure 2D), and BUN (p < 0.001, n = 5) (Figure 2E) in the 1 × 108KP group were higher than those in the control group. Mice in the 1 × 106 KP and 1 × 108 KP groups exhibited closed eyes, increased periocular secretions, chills, and reduced responsiveness to external stimuli from day 1 to day 3. Additionally, the mice tended to huddle together and showed decreased activity. The body weights of the mice in both groups significantly decreased on the third day after intranasal instillation of KP (p < 0.0001) (Figure 2F).
On day 7 after intranasal instillation, flow cytometry analysis showed that the lymphocyte percentage in the 1 × 108 KP group was significantly decreased (p < 0.0001) (Figure 3A) and the proportion of MHC‑II expression in monocytes in the 1 × 108 KP group was significantly decreased compared to the control group (p < 0.05) (Figure 3B).
Day 7 post-intranasal instillation, the MFI of IL-10 and TGF-β1 in CD4+ T cells in the 1 × 108 KP group was significantly increased (p < 0.05) (Figure 4A,C). At 24 h after the secondary infection with PA, the bacterial load in the BALF of the mice in the 1 × 108 KP group was significantly increased (Figure 4B). Within 7 days after intranasal instillation, the mortality rates of the mice in the 1 × 106 KP group and 1 × 108 KP group were significantly increased (p < 0.0001) (Figure 4D). Within 7 days after the secondary infection with PA, the mortality rates of the mice in the 1 × 106 KP group and 1 × 108KP group were significantly increased (p < 0.0001) (Figure 4D).
In conclusion, high-dose KP infection leads to systemic inflammation, multiple organ injury and poor general health in mice. Meanwhile, it further induces immunosuppression, increases host susceptibility to secondary PA infection, and elevates mouse mortality.

Figure 1: Intranasal instillation of Klebsiella pneumoniae. (A) Preparation of mice, 8–10-week-old male C57BL/6 mice. (B) Intraperitoneal injection of 2 µL/g pentobarbital sodium for anesthesia. (C) The mouse's neck was gently held with the thumb placed on the front of the neck, and the rest of the fingers gently grasping the mouse's body, with the mouse's head tilted backward. (D) Slow instillation of the bacterial suspension into the mouse's nostrils using a pipette. (E) After observing that the mouse had inhaled the suspension, the mouse was quickly lifted upward. (F) Slowly put the hands back to their original position. Please click here to view a larger version of this figure.

Figure 2: Inflammatory indicators, organ damage indicators, and 7-day body weight changes in mice. (A) IL-6 level at 12 h. (B) AST level at 24 h. (C) ALT level at 24 h. (D) CK level at 24 h. (E) BUN level at 24 h. (F) 7-day body weight changes in mice. Please click here to view a larger version of this figure.

Figure 3: 7-day lymphocyte and MHC-II flow cytometry. (A) 7-day percentages of lymphocytes and leukocytes in mice. (B) 7-day the proportion of MHC‑II expression in monocytes. Please click here to view a larger version of this figure.

Figure 4: The MFI of IL-10 and TGF-β1 in CD4+ T cells, survival analysis after initial and secondary infection, and BALF bacterial load after secondary infection. (A) MFI of IL-10 in CD4+ T cells. (B) 24-h bacterial count in BALF after secondary challenge with PA. (C) MFI of TGF-β1 in CD4+ T cells. (D) Survival analysis. Please click here to view a larger version of this figure.
Sepsis is a complex clinical syndrome, and the characteristic immunosuppressive state during its late stage is a key factor leading to patient death12,13. Therefore, preclinical models that accurately simulate the pathophysiology of sepsis-induced immunosuppression are crucial for in-depth exploration of its mechanisms and the development of new therapies. In this study, we successfully established and validated a mouse sepsis model induced by intranasal instillation of KP, with a core feature of stably replicating the immunosuppressive state that is highly relevant to clinical practice.
To ensure the reliability and reproducibility of the model, we adopted a series of standardized designs in the establishment process. The standard highly virulent strain Klebsiella pneumoniae ATCC 43816 (KPPR1)14,15 was selected for this protocol. The virulence characteristics of this strain have been extensively validated, enabling stable induction of infection-related pathological changes and laying the foundation for model reproducibility. The standardized procedural design for nasal instillation is the core of ensuring consistent infection routes. To address potential issues such as coughing or sigh-like breathing during the procedure, we established a "rest-observe-stabilize-continue" response protocol, ensuring complete absorption of the 50 µL bacterial suspension and effectively reducing variability in individual infection intensity caused by procedural differences.
The primary advantage of this model lies in its adoption of a natural respiratory infection route, which is more consistent with clinical practice compared with traditional models5,16. Clinically, pneumonia is the most common cause of sepsis6,17, and intranasal instillation directly simulates the clinical pathogenesis of pneumonia-induced sepsis: infection initiates in local lung tissue, then progresses to trigger a systemic inflammatory response and subsequent organ damage. This is consistent with the early elevation of inflammatory factors (IL-6) and changes in liver (ALT, AST), muscle (CK), and renal (BUN) function markers observed in the model, which confirm the occurrence of sepsis-related multiple organ dysfunction18,19, indicating this model exhibits a high degree of similarity to human sepsis in terms of etiology and early pathological progression.
Multiple lines of evidence confirmed that the models established with high concentrations of KP can induce significant immunosuppression. After the acute infection phase, surviving mice showed a significant reduction in the lymphocyte ratio and the proportion of MHC‑II expression in monocytes, which suggests potential reprogramming or exhaustion of the adaptive immune system and represents a classic cytological manifestation of the immunosuppressive state20,21,22,23,24,25,26,27,28. Additionally, the significantly increased MFI of IL-10 and TGF-β1 in CD4+ T cells further confirms that the mice were in an immunosuppressive state26,28,29. One week after the initial infection, the mice exhibited impaired clearance ability and increased mortality following secondary pulmonary infection with a low dose of Pseudomonas aeruginosa. This directly simulates the clinical dilemma where sepsis survivors are susceptible to death from secondary infections30. These results demonstrate that the immunosuppression induced by this model is functionally significant, i.e., the host's antibacterial defense capacity is indeed severely impaired. It should be noted that previous studies have demonstrated that coinfection with PA can promote the proliferation and dissemination of KP31. In this article, we primarily aimed to establish and validate an immunosuppressive model induced by KP infection, using secondary PA challenge to trigger and exacerbate the immunosuppressive state.
Overall, this model provides a standardized animal model for investigating the mechanisms of sepsis-induced immunosuppression. By comparing the model establishment efficiency at different bacterial concentrations, we not only determined the optimal conditions for generating a stable immunosuppressive model but also laid the foundation for further exploration of the dose–response relationship between sepsis severity and the degree of immunosuppression. Of note, this model was established using a single bacterial strain and C57BL/6 mice; thus, its broad applicability to other bacterial strains, other pathogens, or different host genetic backgrounds should be regarded as a potential future direction rather than a confirmed conclusion. This model is expected to promote in-depth research on the molecular mechanisms of sepsis-induced immunosuppression and offer potential directions for the development of novel therapeutic strategies.
The authors declare no conflict of interest.
This work was supported by the National Natural Science Foundation of China (82222038 and 81970259), the Outstanding Young Talents of National Defense Biotechnology (2023-JCJQ-ZQ-001 and 01-SWKJYCJJ06), and the Chongqing Science Fund for Distinguished Young Scholars (CSTB2022NSCQ-JQX0017).
| Name | Company | Catalog Number | Comments |
|---|---|---|---|
| Biochemical incubator | YSEI | SHH-150L | Bacterial incubation |
| C57BL/6J mice | Vital River Laboratories | 8-10 weeks, male | |
| CD116b | BioLegend | 101257 | Flow Antibodies (Anti-Mouse) |
| CD25 | BD bioscience | 740447 | Flow Antibodies (Anti-Mouse) |
| CD3 | BioLegend | 100203 | Flow Antibodies (Anti-Mouse) |
| CD4 | BioLegend | 100544 | Flow Antibodies (Anti-Mouse) |
| CD45 | BioLegend | 103116 | Flow Antibodies (Anti-Mouse) |
| CD8 | BioLegend | 100752 | Flow Antibodies (Anti-Mouse) |
| Cell stain buffer | BioLegend | 420201 | |
| Constant temperature shaker | Honour | HNY-100B | Bacterial cultivation |
| Flow cytometer | Beckman Coulter | CytoFLEX | Flow cytometry analysis |
| High-speed centrifuge | Beckman Coulter | AIIEGRA | |
| IDEXX catalyst one | IDEXX | Catalyst One | Biochemical analysis |
| IL-10 | BioLegend | 505026 | Flow Antibodies (Anti-Mouse) |
| Insulin cyringe | WEGO | 1mL | Blood collection |
| Klebsiella pneumoniae ATCC 43816 | biobw.org | bio-81659 | |
| LB Broth (powder) | Solarbio | L1010 | LB liquid culture medium |
| LB Broth (powder) | Solarbio | L1015 | LB agar plate |
| LY6C | BioLegend | 128010 | Flow Antibodies (Anti-Mouse) |
| Ly6G | BioLegend | 127606 | Flow Antibodies (Anti-Mouse) |
| MHC-II | BioLegend | 107622 | Flow Antibodies (Anti-Mouse) |
| Micropipette | Eppendorf | P100 | |
| Red blood cell lysis buffer | Solarbio | R1010 | |
| Spectrophotometer | Eppendorf | Eppendorf6131 | Optical density measurement |
| Statistical software | GraphPad Software | Graphpad Prism10.0 | |
| Sterile and enzyme-free pipette tips | Biosharp | 200μL | |
| TF diluent buffer | BD Pharmingen | 51-90008101 | |
| TF fix/perm buffer | BD Pharmingen | 51-90008100 | |
| TF perm/wash buffer | BD Pharmingen | 51-90008102 | |
| TGF-β1 | BioLegend | 141406 | Flow Antibodies (Anti-Mouse) |
| Varioskan LUX multimodemicroplate reader | Thermo Scientific | Varioskan LUX |
Request permission to reuse the text or figures of this JoVE article
Request Permission