Establishment and Evaluation of a "Two-Hit" Mouse Model of Sepsis

In recent years, researchers have turned their attention to the study of the mechanism of sepsis-induced immuneparalysis, but there is still a lack of appropriate animal models that reect the process of sepsis-induced immuneparalysis. The purpose of this study was to explore and evaluate whether a "two-hit" model of sepsis can be used as an appropriate animal model to study the mechanism of sepsis-induced immunoparalysis. Firstly, we established a sepsis model in C57BL/6J mice via cecal ligation and puncture(CLP). Routine blood tests, serum ALT and PCT, etc. were analyzed to evaluate the establishment of a CLP sepsis model on day 1 post-CLP. Secondly, the surviving mice were treated with 40µL suspension of P.aeruginosa (Pa) under anesthetic on day 4 post-CLP to establish a "two-hit" model. Gross lung specimens, and pathological examination of the lung tissue, etc. were used to evaluate whether the "two-hit" model was successfully established on days 5 and 11 post-CLP. Finally, the level of serum TNF-α and IL-6, pathological examination of the spleen and the CD4+/CD8 + ratio in blood were detected to evaluate the immune status of the model mice. The above test results proved that the "two-hit" sepsis mouse model was successfully constructed. TNF-α, IL-6, pathological examination of the spleen and the CD4+/CD8 + ratio in blood conrmed that the model was in a state of immunoparalysis. We conclude that this "two-hit" sepsis mouse model is an appropriate animal model, which can successfully replicate the pathophysiological process of in clinical patients with sepsis-induced immunoparalysis.


Introduction
Sepsis is de ned as a life-threatening organ dysfunction caused by a dysregulated host response to infection [1], and is a complex disease with both high morbidity and mortality. Sepsis is much more dangerous than the infection itself, and there are over 18 million cases of severe sepsis worldwide each year. Although most sepsis patients can survive in the acute phase due to the progress made regarding ICU support treatment, sepsis-induced immunoparalysis leading to secondary infection is becoming a leading cause of death [2] .
Recent research has shown that the peak time of sepsis-induced death can be divided into three stages.
The early peak is caused by the in ammatory response in the initial days of sepsis and a late peak after several weeks that is due to organ damage or immunoparalysis. The third upswing occurs after 60-90 days due to persistent immune dysfunction and in ammation in patients with other comorbidities or advanced age [3].
The de nition of Sepsis 3.0 does not clearly de ne the nature and mechanism of the disordered host response to infection, which indicates that researchers continue to have an insu cient understanding of the pathophysiological mechanism of the pathogenesis of advanced immunoparalysis in sepsis [4]. The mechanism currently accepted by researchers is that innateand adaptive immune responses of the host are involved in the pathogenesis of sepsis [5]. The early activation of the innate immune response is the rst line of defense against infection and plays a central role in the initiation process of adaptive immunity. After the onset of sepsis, a strong pro-in ammatory response accompanied by an antiin ammatory response will generally last for several days [5], limiting in ammation and protecting the host. However, if such a compensatory response prevails, it will cause a large amount of death of in ammatory mediators and immune cells in the body. This leads to a more lasting disease state of the host, which is likely to cause secondary infection and viral reactivation in the host [6]. Although modern medical strategies have improved the short-term outcomes in patients with sepsis, they have also led to more persistent diseases tates that shift to an immunoparalysis phenotype, increasing the incidence of delayed death. For this reason, immunoparalysis caused by mid-to-late stage sepsis has become an area of keen interest in many laboratories around the world [3].
Therefore, there is an urgent need to study the pathogenesis of mid-to-late stage sepsis and identify more effective treatment methods; however, there remains a lack of appropriate animal models that can re ect the process of mid-to-late stage sepsis-induced immunoparalysis. In this study, we explored a dual model, termed a "two-hit" model, to evaluate whether it can be used to study the pathogenic mechanism of immunoparalysis in patients with mid-to-late stage sepsis.

Materials And Methods
Animals C57BL/6J mice aged 10-12 weeks old and weighing (24g-26g), were purchased from the Animal Experimental Center of Chongqing Medical University. The mice were housed under speci c pathogenfree conditions at least one week before the experiments, with a 12 h dark/light cycle. All animal experimental operations were performed in accordance with Chongqing Management Approach of Laboratory Animal (Chongqing government order NO.195).

Selection of an Appropriate Cecal Ligation and Puncture (CLP) Sepsis Model
The mice were divided into the following four groups based on the length of cecal ligation: 3/4CLP,1/2CLP, 1/3CLP, and Sham group (n = 10 in each group).The survival rate of the four groups was observed on the fourth day after modeling.
The mice were anesthetized with iso urane (2% 3%) and sterilized with iodophor after preparing the skin. Along the middle of the abdominal white line, the skin of the mice was cut approximately 0.8 cm long, the muscle layer was separated, and the abdominal cavity was explored to expose the cecum (the cecum was primarily located in the left lower abdomen of the mice, and damage to the cecum and blood vessels was avoided during exploration). The intestinal content was gently squeezed with forceps to ll the distal cecum, which was then ligated with a 3/0 silk suture. A 21-gauge needle was used to puncture the cecum at the distal end of the ligation (to avoid the blood vessels), which results in a cecal stula, squeezing out some of the intestinal content, and forming a 1 mm fecal column. Finally, the cecum was carefully placed in the peritoneal cavity, and the abdomen was sutured layer by layer. The 3/4CLP group referred to the distal third-fourth of cecum ligation; the 1/2 CLP group referred to the distal one-half of cecum ligation; the 1/3 CLP group referred to the distal one-third of cecum ligation ( Figure 1A); the Sham group only underwent laparotomy, without distal cecum ligation and puncture. After the operation, all mice were subcutaneously injected with 1 mL of saline and were placed back into the cage after waking up, without water and food restriction( Figure 1B). To control pain, buprenorphine s.c.was injected at a level of 0.05 mg/kg 30 min before operation and thereafter was repeated every 8 h. In the hands of skilled researchers, the CLP procedure can be done about 15-20 min for each animal. Additional time is required for postoperative care and specimen and data collection. According to the mortality rate of these groups, 1/3 CLP operation was selected for all subsequent experiments ( Figure 2A).

Selection of an Appropriate "two-hit" Model of Sepsis
The standard strain of Pseudomonas aeruginosa (P. aeruginosa)(ATCC 27853) was donated by the Department of Microbiology, the Second A liated Hospital of Chongqing Medical University. P. aeruginosa were grown overnight in Nalidixic Acid Cetrimide Broth Medium (Qingdao Rishui Biotechnologies Co., Ltd., Qingdao, China)at 37°C with constant shaking to the stationary phase. Different Pa inoculum densities were prepared using sterilized saline.
At 4 days post-CLP, the surviving mice were divided into ve groups (n = 10) and an esthetized with ether inhalation and raised vertically ( Figure 1C). Next, 10 μL of different P. aeruginosa suspensions (3.0×10 9 , 2.5×10 9 , 2.0×10 9 , and 1.0×10 9 colony-forming unit (CFU/mL) was slowly dripped from the left nasal cavity using a pipette, and the operation was repeated four times. It was necessary to observe whether there were bubbles in the nasal cavity of the mice after each nasal drip to ensure that the bacterial liquid was inhaled into the lungs ( Figure 1D). The control group was treated identically except that sterile saline was in stilled intranasally instead of the bacterial suspension. Survival was recorded for 11 days after CLP. Based on these mortality rates, 2.0×10 9 CFU/mL was selected for all subsequent experiments ( Figure 3A).

Study Design
C57BL/6J mice were placed into one of four groups: 1) CLP + Pa group received CLP followed by intranasal P. aeruginosa at four days post-surgery; 2) the CLP + NS group received CLP followed by an intranasal administration of saline; 3) the Sham + Pa group received sham surgery followed by an intranasal administration of P. aeruginosa; and 4) the Sham + NS group received sham surgery followed by intranasal saline.
The weight of the mice was recorded, and they were sacri ced at 1, 4, 5, and 11 days post-CLP (n =8 mice/time point). The EDTA anticoagulation tube and disposable coagulation vacuum tube were used to collect blood. The blood in the disposable coagulation vacuum tube was placed at room temperature for 30 min, and centrifuged at 1000´ g at 4°C for 10 min, and the serum was stored at -80°C for later use. After euthanasia, the lungs, spleens, and ceca were removed. The lungs and spleens were xed in 4% paraformaldehyde and embedded in para n. Tissue sections were stained with H&E.

Routine Blood Test and ALT
Routine blood tests were measured by the Department of Laboratory Medicine, the Second A liated Hospital of Chongqing Medical University. The level of serum ALT on days 1 and 4 post-CLP were measured using an Alanine Aminotransferase Assay Kit (Nanjing Jiancheng bioengineering institute, Nanjing, China).

Bacterial Culture
The blood in an EDTA anticoagulation tube on day1 post-CLP was diluted to 1:100 using sterile saline.
The diluted blood was then evenly smeared onto blood agar plates using a sterile cotton swab. On day 1 post-CLP, the abdominal cavity of the mice was washed with 5 mL sterile saline to collect the abdominal cavity lavage uid. The abdominal cavity lavage uid was diluted to 1:1000 in the sterile saline and evenly smeared onto a blood agar plate. On day 5 post-CLP, the lungs were ushed with 1 mL sterile saline to collect the bronchoalveolar lavage uid. Next, 10 μL of the bronchoalveolar lavage uid was dropped onto the NAC solid medium. All of the above samples were incubated at 37°C for 24 h.

Enzyme-linked Immunosorbent Assay (ELISA) Analysis
The levels of serum IL-6 and TNF-α at all time points were detected using commercial ELISA kits (4A Biotech Co., Ltd, Beijing, China). The level of serum PCT was detected using a commercial ELISA kit (Elabscience Biotechnology Co., Ltd, Wuhan, China) on days 1 and 4 post-CLP. The reagents and methods were operated in accordance with the manufacturer's instructions.

Lung Index and Spleen Index
The removed lungs and spleens were dried with lter paper, weighed, and both the lung index and spleen index were calculated. The following formula was used: lung index = (lung mass/body mass) ×100%, spleen index = (spleen mass/body mass) × 100%.

Flow Cytometry
The percentage of CD4+ and CD8+ in the whole blood was detected by ow cytometry. The blood in the EDTA anticoagulation tube was dropped into erythrocyte lysis buffer (Tiangen Biotech Co., Ltd, Beijing, China) for 1 min at room temperature, and centrifuged at 2000 g for 5 min. The precipitates were collected and resuspended in 50 μL PBS, the antibodies were conjugated with speci c uorescent dyes, CD3-APC-A750 (Becton, Dickinson and Company), CD4-PerCP5.5 (Becton, Dickinson and Company), and CD8-FITC (Becton, Dickinson and Company)were added and incubated at 4°C for 30 min. Next, they were washed in 1 mL PBS and centrifuged at 2000 g for 5 min. The supernatant was discarded and the cells were resuspended in 200 μL PBS. Finally, the samples were analyzed using a CytoFLEX ow cytometer (Beckman Coulter Co., Ltd, America).

Statistical Analysis
GraphPad Prism 8.2.1 software was used for statistical analysis. The survival rate was analyzed using a Log-rank (Mantel-Cox) test. Student's t-tests and ANOVAs were used as appropriate. The results were displayed as the mean ± SD. A value of P < 0.05 was deemed to be statistically signi cant.

Survival Rates and Identi cation Results of the CLP Model of Sepsis
The survival rate of the mice after CLP surgery was closely related to the length of cecal ligation. The four-day survival rates of the mice in the 3/4 CLP, 1/2CLP, 1/3CLP, and Sham groups were 10%, 40%, 70%, and 100%, respectively. As the mortality rates of the 3/4CLP and 1/2CLP group were too high and nonconducive to subsequent experiments, 1/3CLP was selected as the experimental model for all subsequent experiments (Fig.2a).
The routine blood tests on day 1 post-CLP showed that the levels of white blood cells, neutrophils, lymphocytes, and monocytes in the CLP group were 2.25 ± 1.05, 0.47 ± 0.38, 1.56 ± 0.74, and 0.21 ± 0.139 (*10 3 /μL), respectively, which were signi cantly lower than those in the Sham group (P < 0.05). However, there was no signi cant difference in routine blood tests between the two groups on day 4 post-CLP (P > 0.05, Table1).
On day 1 post-CLP, the level of serum ALT in the CLP group was 66.70 ± 36.33 Karu, which was signi cantly higher than that of Sham group (11.63 ± 2.23 karu, P < 0.05); however, there was no signi cant difference between the two groups on day 4 post-CLP (P > 0.05, Fig.2b).
The level of serum PCT in the CLP group was 958.4 ± 535.2 pg/mL and 607.1 ± 247.5 pg/mL on days 1 and 4 post-surgery, compared to only 124.1 ± 40.9 pg/mL and 156.4 ± 10.0 pg/mL in the Sham group. The level of serum PCT in the CLP group was signi cantly higher than that of the Sham group (P < 0.05; Fig.2c), indicating a severe infection in the CLP group.
A large number of colonies with neat white round edges and smooth surfaces were observed in the blood culture dish and peritoneal lavage uid culture dish of the CLP group on day 1 post-CLP; however, no such results were observed in the blood culture dish or peritoneal lavage uid culture dish of the Sham group, and the blood agar medium was bright red (Fig.2d). Therefore, these results demonstrated that the CLP model of sepsis had been successfully established.
Survival Rates and Identi cation Results of the "two-hit" Model of Sepsis According to the survival curve of the "two hit" model, on day 11 following CLP, four mice in the CLP + Pa group died, one mouse in the Sham + Pa group died, and all mice in the CLP + NS and Sham + NS groups survived (Fig.3b).  (Table.2). Compared with the CLP + Pa group, the white blood cells and neutrophils were signi cantly decreased in the three groups (P < 0.05).
On day 5 post-surgery, the bronchoalveolar lavage uids of the four groups were cultured. Green colonies were observed in the NAC solid medium in the CLP + Pa group and Sham + Pa group, and the number of colonies in the CLP + Pa group was greater than that in the Sham + Pa group. However, no colonies were observed in either the CLP + NS group or Sham + NS group (Fig.3c).
We further observed the appearance of the lung tissue. On day 5 post-surgery, there were obvious bleeding points in the lung tissues of the CLP + Pa and Sham + Pa groups, especially in the CLP + Pa group. On day 11 post-surgery, the lung tissues of the surviving mice in the CLP+Pa group were consolidated dark red in color, and the majority of the lung lobes were affected, whereas there was only partial lung tissue damage in the Sham+Pa group. The lung tissue of the mice in the CLP+NS and Sham + NS groups was normal at the both time points (Fig.3d).
HE slices of the lungs in the CLP + Pa and Sham + Pa groups showed that the lung tissues exhibited widened alveolar septum, damaged alveolar structure, hemorrhage, serous exudation, and a large amount of in ammatory cell in ltration on day 5 after the operation. The degree of lung tissue injury was more severe in the CLP + Pa group. On day 11 post-surgery, there was greater consolidation of the alveolar cavity, focal pulmonary interstitial hemorrhage, and a large number of in ammatory cell in ltration in the connective tissues of the alveoli and alveolar septum in the CLP + Pa group. In the Sham + Pa group, only part of the in ammatory cells had in ltrated into the lung tissue. In the CLP + NS and Sham + NS groups, the lung tissue HE staining showed no signi cant change (Fig.3e).
On day 5 post-CLP, the lung indexes of the CLP + Pa, CLP + NS, Sham + Pa, and Sham + NS groups were calculated respectively: 0.72 ± 0.05%, 0.606 ± 0.02%, 0.70 ± 0.04%, and 0.51 ± 0.02%.The lung index of CLP + Pa group was signi cantly higher than that of the CLP + NS group and Sham + NS groups (P < 0.01). On day 11 post-CLP, the lung index of the four aforementioned groups were 0.74 ± 0.04%, 0.56 ± 0.02%, 0.66 ± 0.01%, and 0.51 ± 0.02%, respectively. The lung index of the CLP + Pa group was signi cantly higher than that of the other three groups (P < 0.01), indicating a serious lung infection in the CLP + Pa group (Fig.3f). Together these results prove that the "two-hit" model of sepsis was completely constructed.
The Body Weight Results and Abdominal Organ Changes of the "two-hit" Model The body weight changes of the "two-hit" mice were continuously recorded within 11 days post-CLP. The results showed that the weight of the mice in the CLP operation group decreased continuously within 5 days after the operation, and increased slowly. The weight of the mice in the Sham operation group temporarily decreased on the rst day after operation and subsequently underwent a gradual increase. The difference between the two groups was statistically signi cant (P < 0.01). After an intranasal instillation of P. aeruginosa, the weight of the mice in the Sham + Pa group decreased for a short time and gradually increased. On day 11 post-CLP, the weight gain of the mice in the CLP + Pa group was signi cantly lower than that in the other three groups(P < 0.01), which suggests that the infection in the CLP + Pa group was more serious than that of the other three groups (Fig.4a).
In both the CLP + Pa and CLP + NS groups, pale necrosis occurred at the distal end of cecal ligation on day 1 post-CLP; white purulent necrosis, adhesion with surrounding tissues, and edema appeared on day 4; both purulent secretion and the area of abdominal adhesion increased on day 5; and the necrotic area was wrapped and could not be completely separated on day 11. However, in both of the Sham + Pa and Sham + NS groups, the shape and structure of the cecum were intact, and no abnormalities were present throughout the experiment (Fig.4b).
The changes in the spleen size of each of the four groups were observed throughout the experimental cycle. From day 4 post-CLP, the spleens in the CLP + Pa and CLP + NS groups were signi cantly enlarged. On day 11 post-CLP, the size of the spleen in the CLP + Pa group was substantially larger than that of the other three groups, suggesting the in ammatory reaction in the CLP + Pa group was the most severe (Fig.4c).
On day 1, 4, 5 and 11 post-CLP, the spleen capsule in the CLP + Pa group gradually thickened, multinucleated macrophages gradually increased, and the number of white pulp decreased signi cantly, and the white pulp and red pulp boundary in the CLP + Pa group gradually blurred. The number of white pulp in the CLP + NS group was slightly reduced, and the boundary between the white pulp and the red pulp was still distinguishable. The brous capsule in the Sham + Pa group was thickened, and the ratio of white pulp to red pulp was normal. The capsule of the spleen tissue in the Sham + NS group was smooth, the Shape and size of the white pulp was regular, the boundary between the white and red pulp was clear and there was no abnormality (Fig.4d).
The spleen index was calculated at four different time points. The results showed that the spleen index was 0.26 ± 0.03%, 0.524 ± 0.06%, 0.70 ± 0.04%, and 1.36 ± 0.26%, respectively, in the CLP + Pa group; 0.25 ± 0.04%, 0.50 ± 0.03%, 0.67 ± 0.11%, and 0.89 ± 0.06%, respectively, in the CLP + NS group; 0.23 ± 0.02%, 0.29 ± 0.03%, 0.32 ± 0.03%, and 0.45 ± 0.03%, respectively, in the Sham + Pa group; and 0.24 ± 0.05%, 0.27 ± 0.05%, 0.29 ± 0.05%, and 0.28 ± 0.04%, respectively, in the Sham + NS group. The spleen index of the mice in the CLP operation group was signi cantly higher than that in Sham operation group(P < 0.01). However, on day 11 after CLP, the spleen index of the mice in the CLP + Pa group also signi cantly differed from that in the CLP + NS group(P < 0.01), which further highlighted the severity of the infection in the CLP + Pa group (Fig.4e). Our results showed that the spleens were in a state of hypersplenism due to infections and the immune cells were continuously consumed.
Immunization Results of the "two-hit" Model On day 11 post-CLP, the white blood cells in the CLP + Pa group were 31.74 ± 7.14*10 3 /μL, which was signi cantly higher than that in the CLP + NS group , Sham + Pa group and Sham + NS group (P < 0.05 or P < 0.01); the monocytes were 3.11 ± 1.28*10 3 /μL, which was signi cantly higher than that in the CLP + NS and Sham + NS groups (P < 0.05). However, since the number of neutrophils and lymphocytes in the CLP + Pa group were exceeded the detection range, statistical analysis could not be performed (Table 3).
We further observed the changes in the level of TNF-α and IL-6. On day 1 post-CLP, the level of serum TNF-α was 167.00 ± 42.68 pg/mL in the CLP group and 45.90 ± 0.38pg/mL in the Sham group (P < 0.01), indicating that TNF-α was activated following CLP. On day 5 post-CLP, the level of serum TNF-α in the CLP + Pa and Sham + Pa groups were 110.69 ± 9.18pg/mL and 148.43 ± 13.42pg/mL, respectively. These ndings were signi cantly different from those in the CLP + NS group (45.90 ± 0.38pg/mL) and Sham + NS group (45.81 ± 0.77pg/mL) (P < 0.0001), which suggests that serum TNF-α was produced after an intranasal instillation of P. aeruginosa. The level of serum TNF-α in the CLP + Pa group was signi cantly lower than that in the Sham + Pa group (P < 0.01), suggesting that the CLP operation may inhibit TNF-α reactivation. However, no signi cant difference was observed regarding the level of serum TNF-α between the four groups on days 4 and 11 post-CLP (Fig.5a). Moreover, the change trend of IL-6 was similar to that of TNF-α (Fig.5b).
The ratio of CD4+/CD8+ in whole blood was also detected at four different time points by ow cytometry to observe the cellular immune function. The results showed that the ratio of CD4+/CD8+ in the CLP + Pa group was 0.57 ± 0.10 on day 11 post-CLP, which was signi cantly lower than that of the other three groups (the ratio of CD4+/CD8+ in the CLP + NS group, Sham + Pa group, and Sham + NS group was 0.91 ± 0.04, 1.17 ± 0.03, and 1.48 ± 0.04, respectively) (P < 0.05 or P < 0.01, Fig.5c,d). These ndings indicate that the "two-hit" model of sepsis may lead to the the decreased presence of immune cells . The above results indicate that the "two-hit" model of sepsis could lead to immunoparalysis.

Discussion
As the name suggests, the "two-hit" animal model simulates an attack by two pathogens on the host, induced by CLP surgery and secondary infection. The CLP model of sepsis mainly simulates clinical sepsis induced by acute peritonitis, which currently represents the gold standard in animal models for studying sepsis [7]. The course of sepsis can pass through different stages, which are clearly de ned by human physiological parameters. However, no physiological monitoring exists in animal models and the severity of sepsis can only be determined by mortality and time of death [8]. In our experimental model, we reproduced three CLP models by selecting different lengths of cecal ligation. By observing the survival rate and general situation of the mice at four days post-CLP, 1/3 cecal ligation was selected as the experimental model as it caused mild sepsis.
Cytokine storm syndrome is induced by innate immune activation during the early stage of sepsis, which may lead to multiple organ failure and early death [9]. Procalcitonin (PCT) is a calcitonin-promoting hormone that is commonly used as a clinical infection index. The level of serum PCT in sepsis patients is signi cantly increased, and can be detected within 24 h post-infection, peaks at 12-24 h, and is closely related to the severity of infection [10]. PCT was detected in our mouse model of sepsis, which is in accordance with the response of clinical patients. TNF-α and IL-6 are the important mediators of the acute phase of the in ammatory response. TNF-α expression is closely related to the severity of sepsis, which can be used to predict the severity of sepsis [11]. IL-6 is an important pro-in ammatory cytokine in the early stage of in ammation, which increases in blood circulation after infection and trauma [12]. Our ndings showed that the levels of PCT, TNF-α, and IL-6 were signi cantly increased on day 1 post-CLP, which con rmed the presence of a cytokine storm syndrome. Moreover, the level of serum ALT also displayed a signi cant increase, which coincided with acute organ dysfunction. This phenomenon is consistent with the clinical observation that sepsis can cause multiple organ dysfunction.
At present, clinical studies have con rmed that patients with acute sepsis have a high probability of death due to secondary infection after the rst three days or longer after the onset of sepsis [13]. One of the main causes of death is secondary infection caused by sepsis-induced immune dysfunction, whereas pneumonia is a common secondary infection [9]. P.aeruginosa is the most common multidrug-resistant gram-negative bacteria that can cause ventilator-associated pneumonia(VAP) and nosocomial pneumonia [9]. In addition, compared with other types of pneumonia, patients with P. aeruginosa pneumonia are more prone to multiple organ failure and death [14][15]. Therefore, P. aeruginosa represents the best choice as a "two-hit" sepsis model strain.
It has been reported that the immune dysfunction turning point for sepsis induced by CLP may occur on day 4 post-CLP [16]. Our study also con rmed that CLP mice no longer died from day 4 post-CLP. Thus, we chose to administer the P. aeruginosa suspension via a "nasal drip" infection at day 4 post-CLP as a second hit to observe the general condition and survival rate of the mice. The bronchoalveolar lavage on day 5 post-CLP and the survival rate of the "two-hit" model revealed that this model could successfully simulate sepsis patients with secondary infection.
As we all know, neutrophils mainly play the role of the rst line of defense in in ammation sites, such as phagocytosis, degranulation and formation of reactive oxygen species [17]. Routines blood tests of the "two-hit" model showed that the number of neutrophils gradually increased from day 5 post-CLP, and it was more than three times of that of the normal mice on day 11 post-CLP. Combined with manifestation of hypersplenism in mice, we thought it was a leukemoid reaction caused by severe infection. Leukemoid reaction refers to the continuous increase of neutrophils, more than 50000 cells per microliter, accompanied by the increase of neutrophil precursors when the cause of leukemia is excluded.The main cause of leukemoid reaction are severe infection, poisoning, malignant tumor, severe hemorrhage or acute hemolysis [18]. However, there is no diagnostic criteria for leukemia reaction of murine.
The derangements in the in ammatory response and a state of adaptive immunoparalysis, which are characterized by the decrease of primary humoral immune response [19]. On day 5 post-CLP, the levels of TNF-α and IL-6 in the CLP + Pa group were signi cantly higher than those in the CLP+NS and Sham + NS groups, but lower than those in the Sham + Pa group. The phenomenon is caused by excessive consumption of in ammatory factors in humoral immune dysfunction after CLP.
Splenic pulp includes white pulp(WP,cells with weakly stained nuclei), red pulp(RP) and the marginal zone(MZ).The RP is composed of blood vessels, lining endothelial cells, and macrophages. These macrophages screen and remove pathogens and soluble antigens from the systemic circulation. WP represents lymphoid tissue, which is composed of central arterioles, lymphocytes and antigen-presenting cells (mainly dendritic cells and macrophages).WP produces speci c immune responses to antigens [20].
The pathological section of mice found that the white pulp in the CLP+Pa group was signi cantly reduced compared with the other three groups, which indicated that the immune function of the "two-hit" model has been compromised.
Patients who survive the acute phase of sepsis show long-term damage to immune function due to the decrease in the number and function of many immune cell populations [21]. With the increasing use of antibiotics and various anti-in ammatory factors in recent years, most patients will survive the systemic acute in ammatory reaction period, but subsequently enter a longer-lasting immune dysfunction state [22]. Studies have found that the key cause of immune dysfunction in patients with sepsis is the death of lymphocytes, especially T and B lymphocyte apoptosis, is a major factor in immune dysfunction [23]. CD4+ can help B lymphocytes produce antibodies and induce T lymphocytes to transform into effector cells, and is a surface marker of Ti/Th. CD8 + mediates the destruction and lysis of host cells after pathogen infection, and is a protein in Ts. In mammals, the ratio of CD4 +/CD8 + is normally> 1, depending on the species [24]. And the CD4+/CD8+ ratio is highly correlated with host immune function.
An increase in CD4 positive cells and an increase in the CD4/CD8 ratio indicate a strong immune system [25]. Therefore, by continuously monitoring the ratio of CD4+ and CD8+ T cells in the whole blood, we found that the CD4+/CD8+ ratio in the mice in the "two-hit" model was signi cantly lower than that of the other three groups on day 11 post-surgery, which was consistent with the immune dysfunction status of patients with advanced sepsis.
However, there are limitations associated with the "two-hit" model. First, the impact of cecal puncture on humans is a fatal blow but the impact on rodents is not obvious [26]. Secondly, the severity of sepsis caused by CLP is related to multiple factors, including the length of cecal ligation, the puncture needle size, and the number of perforations. Of course, the stability of the model is also closely related to the gender and age of the mice. Thirdly, the "two hits" are administered by "nasal drip"; however, in the process of administering the "nasal drip", the bacterial solution may be swallowed by the mice rather than inhaled, which may cause different degrees of infection. Fourth, sepsis patients are often accompanied with endocrine, tumor, and other diseases in the clinic. The establishment of a "two-hit" model with only healthy mice could not cover all clinical conditions. Finally, a mouse model cannot completely simulate the changes in the whole blood and dynamic changes in the in ammatory factors in patients with sepsis.

Conclusions
This "two-hit" model focuses on replicating the process of clinical sepsis patients entering the immune dysfunction stage during the early stages of sepsis. The immune function of the mice was evaluated by the continuous observation of routine blood tests, in ammatory factors, and the CD4+/CD8+ ratio.
Together, these ndings demonstrate that this "two-hit" model can be used as an appropriate animal model to study the mechanism of immune dysfunction in sepsis and the associated drug development.  The detailed process of establishing the "two-hit"sepsis mouse model. a.Schematic illustration of characterized positions of cecal ligation to induce 3/4 CLP,1/2 CLP and 1/3 CLP in mice. b. CLP and Sham operations were performed. A median laparotomy of 0.8 cm was performed and the cecum was exposed. The distal one-third of the cecum was ligated with 3/0 silk suture and punctured once with a 21gaugeneedle. The cecum was gently squeezed to extrude some feces before returning it to the abdomen. The Sham group only underwent a laparotomy, without distal cecum ligation and puncture.The abdominal wall and skin were closed in two layers and 1mL saline was administered intraperitoneally after surgery. c.The surviving mice on day 4 post-CLP were anesthetized with ether inhalation. d. The mice were infected with Pseudomonas aeruginosa via "nasal drops" under anesthesia.  and 400×. e.The spleen index of the "two-hit" model at different time points post-CLP. Compared with the CLP+Pa group, ** P < 0.01; *** P < 0.001; **** P < 0.0001.

Figure 5
Immunization damage in the "two-hit" model mice. Changes in the level of serum cytokines TNF-α (a),IL-6 (b), and whole blood CD4+/CD8+ ratio (c) in mice that received the "two-hit" model at different time points post-CLP. The levels of TNF-α and IL-6 were determined by an ELISA. (d) The percentage of CD4+ and CD8+ in the whole blood were detected by ow cytometry. Data are presented as the mean±SD (n = 8 mice per group). Compared with the CLP + Pa group, * P < 0.05; ** P < 0.01; **** P < 0.0001.