Introduction

Bacterial meningitis, including Escherichia coli meningitis, is a severe inflammatory disease of the central nervous system (CNS) with elevated mortality and morbidity rates in children. From 30 to 50 percent of survivors continue to suffer neurological sequelae despite treatment with effective antibiotics, adjuvant corticosteroids and critical care1,2. Blood–brain barrier (BBB) disruption is considered to be a major contributor to the neurological sequelae, which is a complex component that prevents pathogen from invading the CNS3,4. Neuroprotective agents may significantly reduce neurological complications and sequelae in bacterial meningitis. Unfortunately, there have been relatively few drugs that protect BBB.

Ulinastatin consists of 143 amino acid residues and two Kunitz-type protease-inhibitor domains arranged in tandem to inhibit serine proteases, granulocyte elastase, and numerous other enzymes. Therefore, it has been shown to reduce inflammatory responses and improve tissue perfusion1,5,6,7,8,9,10,11. Ulinastatin has been widely used in the treatment of acute severe pancreatitis, sepsis, acute respiratory distress syndrome (ARDS), systemic inflammatory response syndrome (SIRS), and multiple organ dysfunction syndrome (MODS). Moreover, recent studies have suggested that ulinastatin had neuroprotective effect on rats with cerebral trauma, hepatic encephalopathy, and ischemia/reperfusion injuries by reducing BBB permeability12,13,14. The tight junction protein zonula occludens-1 (ZO-1) is a major component of BBB. ZO-1 degradation induced by ischemia–reperfusion injury in rats can be ameliorated by ulinastatin. Rats with bacterial meningitis showed significant decrease in ZO-1 expression, resulting in increased vascular permeability, cerebral edema, and consequently cerebral cell death15. Phosphorylation of ZO-1 on serine, threonine, and tyrosine residues decreases transmembrane resistance and increases barrier permeability16. Protein kinase C (PKC) is a class of serine/threonine specific kinase family, which can catalyze the phosphorylation of serine or threonine residues on various protein substrates. Ulinastatin is a serine proteases inhibitor as well as may alleviate ZO-1 loss17. Therefore, we hypothesized that ulinastatin had neuroprotection in bacterial meningitis via inhibiting PKCα phosphorylation and reducing ZO-1 loss.

Materials and methods

Animals and grouping

This study was approved by the the Medical Ethics Committee of Xiamen Children’s Hospital (ethics numberXMCH2021-187). All methods were carried out in accordance with relevant regulations and ARRIVE guidelines. Three weeks old healthy male Sprague–Dawley (SD) rats (n = 40, 100–120 g) were purchased from Shanghai Slack Laboratory Animal Co. Ltd. (Shanghai, China). The rats were bred under a 12 h light/12 h dark cycle with an ad libitum access to food and water.

Forty healthy male SD rats were randomly divided into five groups, which were subjected to the following interventions: (1) Control group (n = 8): no intervention. (2) E. coli group (n = 8): the E.coli meningitis model was established by a cerebellar medulla cistern puncture18 50 ul of CSF was extracted and 50 ul of E. coli suspension was injected in the same way, immediately followed by an injection of 0.5 ml of saline in the tail vein; (3) E. coli + UTI group (n = 8): E. coli rat meningitis model was established, and UTI (50,000 U/kg,0.25ml) was injected through the tail vein, immediately followed by an injection of 0.25 ml of saline in the tail vein; (4) E. coli + UTI + PMA group (n = 8): E. coli rat meningitis model was established, and UTI (50,000 U/kg,0.25 ml) was injected through the tail vein, immediately followed by an injection of PMA (200 ug/kg,0.25ml)through the tail vein. (5) E. coli + PMA group (n = 8): E.coli meningitis of model rats was established, and PMA (200 ug/kg 0.25 ml) was injected through the tail vein, immediately followed by an injection of 0.25 ml of saline in the tail vein.

Bacteria and drugs

E. coli K1 was obtained from the Laboratory Department of Zhongshan Hospital Affiliated to Xiamen University (Xiamen, China). We inoculated Escherichia coli into Miller LB (Luria–Bertani) broth, placed it in a bacterial incubator at 37 °C, and stored it at− 80 °C for later use. E.coli strains were inoculated in LB medium, grown overnight at 37 ℃ in 5% CO2 incubator, harvested after 24 h of growth, centrifuged, rinsed with normal saline, repeated twice, centrifuged again, and diluted with saline to 106 cfu/ml. Ulinastatin (UTI) was purchased from Techpool Biopharma Co. Ltd. (Guangdong, China, 031909133). ZO-1 antibody was obtained from Santa Cruz Biotechnology, Inc., USA, (SC-33725), β-actin antibody was produced by Santa Cruz Biotechnology, Inc., USA (SC-4718), p-PKCα antibody was purchased from Affinity Biosciences company (Jiangsu, China, AF8396), t-PKCα antibody was purchased from Affinity Biosciences Co.Ltd (Jiangsu, China; AF 6196). PMA was manufactured by Merck KGaA (Darmstadt, Germany, P8139).

Neurological function assessment

Loeffler five-point neurobehavioral score was used to evaluate the neurobehavioral scores of rats at 12 h after modeling. The following point score was applied: 5 points: can hold the back during normal activity, turns over in 5 s successfully. 4 points: reduced voluntary movement, able to turn over within 5 s. 3 points: turning over can be successful but the time is more than 5 s. 2 points: cannot roll over. 1 point: no exercise. 0 points: death.

Hematoxylin and eosin (H&E) staining

The rats were administered intraperitoneal tribromoethanol (Dalian Meilun Biotechnology Co.LTD, China, MA 04781-1) at a dose of 25 ml/kg, 12 h after modeling, for anesthesia prior to euthanasia. The brain tissue was dehydrated, transparent, embedded, cut into 6-µm-thick sections, baked, dewaxed to water, stained with hematoxylin for 3 min, and rinsed with water. Then, it was stained with eosin for 3 min, rinsed with water, and placed in an oven to dry. Further, the specimen was sealed with neutral resin after drying. The pathological changes in the rat brain were observed under a light microscope.

Nissl staining

The brain sections were dehydrated with 100%, 95.0%, and 70.0% alcohol for 1 min, respectively, and washed with ultrapure water. Then, the tablets were stained with 0.1% tar violet for 30 min and washed with ultrapure water. Next, they were dehydrated with 75.0%, 85.0%, 95.0%, and 100% gradient alcohol (1–2 min each). Finally, xylene was transparent three times, 2 min each time. The slices were sealed with neutral resin, and the number of neurons was counted under a light microscope.

TUNEL staining

The sections were routinely dewaxed to water, and 20 μg/ml DNase-free proteinase K was added by drop. The reaction was carried out at 37 °C for 15–30 min, washed with PBS three times. A volume of 50 μl of TUNEL assay solution was added, and the samples were incubated at 37 ℃ for 60 min in the dark, washed three times with PBS, and sealed. Fluorescence microscopy was used to observe at 100 times field of view (excitation wavelength 450–500 nm, emission wavelength 515–565 nm). Nine fields were randomly selected for each section. Images were taken and optical density analysis was performed using Image Pro Plus 6.0 Image analysis software. The integral optical density (IOD) value of the green fluorescence area was detected, and the average value was taken as the IOD value of each section.

Immunohistochemical staining

Brain tissue sections were routinely dewaxed to water, blocked with H2O2 for 10 min, blocked with donkey serum for 10 min, and incubated with primary antibody at 4 ℃ overnight. Then secondary antibody was added, followed by incubation at 37 ℃ for 30 min, development with DAB for 3 min, and staining with hematoxylin for 3 min. All aforementioned steps were followed by washing with PBS 3 times, 1 min each time. Then, we performed routine dehydration with 95% ethanol, followed by baking and sealing. Each section was observed under a light microscope, and nine fields of view were randomly selected for each section. Image Pro Plus 6.0 Image analysis software was used for optical density analysis after images were taken. The integral optical density (IOD) value of the positive staining area was measured, and its average value was taken as the IOD value of each section.

Western blotting

A homogenate was prepared at a rate of 500 ul per 50 mg of tissue. The tissue was cut into pieces, homogenized, and its protein content was determined by using BCA Protein Assay Kit (Beyotime Biotechnology, China). Further, 2 × SDS loading buffer was added, followed by boiling in water at 100 ℃ for 3 min. SDS-PAGE was prepared for protein solution separation. Then the protein was transferred to a Polyvinylidenfluorid (PVDF) membrane. The diluted solution of primary antibody was prepared, 3% milk-TBST was prepared, and the primary antibody was added. The membrane was incubated at 4 ℃ overnight, and washed with TBST three times, for 5 min each time. Next, the membrane was incubated in 3% milk-TBST-diluted secondary antibody (1:1000) at 37 ℃ for 1.5 h. Then membrane was washed with TBST three times, for 5 min each time. The protein was detected by ECL Kit (Beyotime Biotechnology, China) and pictures were taken by image system. After scanning the bands of each group, the gray levels were measured by ImageJ.

Statistical analysis

Data were statistically compared and analyzed using SPSS 20.0 software. Data were expressed as means ± SD. One-way ANOVA was used for comparison of means between groups, and the LSD test was implemented for pairwise comparison. P < 0.05 was considered to indicate statistically significant differences.

Results

Effect of ulinastatin on Loeffler neurobehavioral score

Loeffler neurobehavioral score was used to assess behavioral changes. The E. coli was cultured in the cerebrospinal fluid of the model groups. The rats in the model groups showed lethargy, poor spirit, less movement, hemiplegia, convulsion, and ate less or no food. The Loeffler neurobehavioral score in the E.coli group was significantly lower than control group(n = 8, P < 0.001), After ulinastatin treatment, the score increased significantly(n = 8, P < 0.001), and PMA attenuated this effect (n = 8, P < 0.001) (Fig. 1).

Fig. 1
figure 1

Loeffler score in rats of each group Loeffler five-point neurobehavioral score was used to evaluate the neurobehavioral scores of rat at 12 h after modeling; Data were expressed as Mean ± SD, (n = 8); **P < 0.01, ***P < 0.001.

Full size image

Effect of ulinastatin on brain histomorphological analysis

Hematoxylin and eosin (H&E) staining was used to visualize the cellular morphology of the meninges and brain tissues. The E.coli group showed typical inflammatory manifestations, in which necrosis of the nerve cells was obvious near the meninges. Meanwhile, the meninges were edematous and congested, with a large number of inflammatory cells. The E. coli + UTI Group had decreased neural necrosis in the cerebral cortex and a slightly lower number of the inflammatory cells under the meninges than the E.coli group. However, The pathological manifestations were aggravated when added PMA, more gray matter necrosis was present in the cerebral cortex, meningeal congestion, swelling, and inflammatory cell infiltration were increased, and even submeningeal cortical hemorrhage had occurred (Fig. 2a). Furthermore, Nissl staining was used to observe neuronal damage. The control group showed a large number of nissl bodies arranged regularly under the microscope, stained nuclei were intact and with normal morphology. The number of nissl bodies in E. coli group decreased (n = 4, P < 0.001), nuclear pyknosis, deformation, lysis, and other morphological abnormalities were observed. After ulinastatin treatment, nissl bodies increased significantly (n = 4, P < 0.01), increased nuclei with normal morphology, and fewer cells with abnormal morphology. Howerer, When combined with PMA, nissl bodies decreased again (n = 4, P < 0.05) (Fig. 2b,c). These findings suggested that ulinastatin reduced meningeal inflammatory infiltration, while PMA can aggravate the progression of inflammation.

Fig. 2
figure 2figure 2

H&E and Nissl staining of brain tissue in rats of each group Prcm medial precentral area, Prcl lateral precentral area, SI sensorimotor cortex I, ACd anterior cingulate cortex, dorsal part, CPu caudate putamen, acb accubens nucleus. (a). H&E staining of brain tissue in rats of each group; Black Up-Pointing Traingle Indicates degeneration, necrosis and lysis of nerve cells; Black Diamond Indicates meningeal congestion and inflammatory cell aggregation; (b). Nissl staining of brain tissue in rats of each group; Black Star Indicates Nissl bodies of nerve cells; (c). Statistics of Nissl bodies count in brain tissue of rats in each group; Data were expressed as Mean ± SD, (n = 4); *P < 0.05, **P < 0.01, ***P < 0.001.

Full size image

Effect of ulinastatin on cerebral cell apoptosis

TdT-mediated dUTP nick end labeling (TUNEL) staining was used to detect cerebral cell apoptosis. It could be seen that compared with the control group, The integral optical density (IOD) value of TUNEL positive staining area in E.coli group rats was significantly increased (n = 4, P < 0.001). However, the IOD value was significantly decreased after ulinastatin treatment (n = 4, P < 0.001). Nevertheless, the IOD value was significantly increased (n = 4, P < 0.001) when PMA were combined to treat E.coli rats. E.coli + PMA group had the most apoptotic cells. These datas suggested that ulinastatin reduced apoptosis, while PMA did the opposite (Fig. 3a,b).

Fig. 3
figure 3

TUNEL staining of brain tissue in rats of each group Prcm medial precentral area, Prcl lateral precentral area, SI sensorimotor cortex I, ACd anterior cingulate cortex,dorsal part, CPu caudate putamen, acb accubens nucleus, TUNEL Terminal Deoxynucleotidyl Transferase Mediated dUTP Nick End Labeling. (a). TUNEL staining of brain tissue in rats of each group; White diamond indicates TUNEL positive staining, it appears as a bright green fluorescence, located in the nucleus, indicating DNA breakage and apoptosis. (b). Statistics of TUNEL positive staining in brain tissue of rats in each group; Data were expressed as Mean ± SD, (n = 4); *P < 0.05, ***P < 0.001.

Full size image

Effect of ulinastatin on expressions of PKCα phosphorylation and ZO-1

Immunohistochemistry and immunofluorescence staining was used to detect the expressions of p-PKCα and ZO-1. The expression of ZO-1 in the E. coli group was significantly reduced compared with control group (n = 4, P < 0.001), but increased significantly after ulinastatin treatment (n = 4, P < 0.001). However, compared with the E.coli + UTI group, the expression of ZO-1 in E.coli + UTI + PMA group was significantly reduced (n = 4, P < 0.001) (Fig. 4a,b). Compared with the control group, the expression of p-PKCα in the E. coli group was significantly increased (n = 4, P < 0.001). The number was significantly decreased in the E.coli + UTI group (n = 4, P < 0.001). PMA aggravated PKCα phosphorylation and attenuated the effect of ulinastatin (n = 4, P < 0.001) (Fig. 5a,b). Western blotting was used to the detect the relative expression levels of ZO-1 and p-PKCα, verifying the above immunohistochemical results (Fig. 6). These datas suggested that ulinastatin may increase ZO-1 expression by inhibiting PKCα phosphorylation.

Fig. 4
figure 4figure 4

IF staining of ZO-1 of brain tissue in rats of each group IF Immunofluorescence; DAPI 4’,6-diamidino-2-phenylindole; (a) IF staining of ZO-1 of brain tissuein rats of each group; DAPI staining shows the nucleus, it appears as a blue fluorescence; White diamond Indicates ZO-1 positive staining, it appears as a bright red fluorescence, located in the cytoplasm of vascular endothelial cell. Merge was merged from DAPI staining and ZO-1IF staining. (b) Statistics of ZO-1 positive staining in brain tissue of rats in each group; Data were expressed as Mean ± SD, (n = 4); ***P < 0.001.

Full size image
Fig. 5
figure 5

IHC staining of p-PKCαof brain tissue in rats of each group Prcm medial precentral area, Prcl lateral precentral area, SI sensorimotor cortex I, ACd anterior cingulate cortex, dorsal part, CPu caudate putamen, acb accubens nucleus; (a) IHC staining of p-PKCα of brain tissue in rats of each group; Black Up-Pointing Traingle Indicates p-PKCα positive staining; p-PKCα positive staining sites were brownish yellow, located near the nucleus in the cytoplasm; (b) Statistics of p-PKCα positive staining in brain tissue of rats in each group; Data were expressed as Mean ± SD, (n = 4); ***P < 0.001.

Full size image
Fig. 6
figure 6

Expression of p-PKCα& ZO-1of brain tissue in rats of each group (a). western blotting of p-PKCα& ZO-1of brain tissue in rats of each group; (b). Statistics of p-PKCαin brain tissue of rats in each group; (c). Statistics of ZO-1in brain tissue of rats in each group; Data were expressed as Mean ± SD, (n = 4); **P < 0.01, ***P < 0.001. Fig. 6 is a representative image of the Western blot results for each group in this experiment. The original Western blot bands and membrane photos for all experimental rats can be found in Supplementary Image S1 online.

Full size image

Discussion

Our findings demonstrated that intravenous ulinastatin administration immediately after E.coli meningitis modeling in rats is neuroprotective, as assessed by neurobehavioral scores, inflammatory infiltration, histomorphological results, and brain cell apoptosis. One possible protective mechanism was that ulinastatin inhibits PKCɑ phosphorylation and thus reduces ZO-1 damage (Fig. 7).

Fig. 7
figure 7

Neuroprotective effects of ulinastatin on Escherichia coli meningitis rats through inhibiting PKC αphosphorylation and reducing zonula occludens-1 degradation.

Full size image

Hematoxylin and eosin (H&E) staining, Nissl staining and TUNEL staining showed that ulinastatin attenuated inflammatory response and neuronal apoptosis in cerebral cells,similar to the results in lung, heart, liver and kidney cells subjected to lipopolysaccharide (LPS)-induced sepsis. LPS is a principal component of the outer membranes of gram-negative bacteria which is known to induce the production of several inflammatory cytokines such as tumor necrosis factor-α (TNF-α), interleukin-1β(IL-1β), IL-6, and IL-2, which recruit and activate neutrophils and lead to increase neutrophil-endothelial adhesion, neutrophil accumulation, transendothelial migration of neutrophil6. Ulinastatin could alleviate these inflammatory cytokines, reduced inflammatory cell infiltration and apoptosis19,20. It inhibited apoptosis via downloading incRNA metastasis-associated lung adenocarcinoma transcript 1 (MALAT1) and Enhancer of Zeste Homolog 2 (EZH2), decreasing pAkt, NF-κB and Bcl-2 in cardiac microvascular endothelial cells (CMVECs)21,22, or blocked High Mobility Group Protein 1(HMGB1)/Toll Like Receptor 4(TLR4) NF-KB signal transduction in kidney23,24. In addition, ulinastatin also has anti-inflammatory and anti-apoptotic effects in the central nervous system (CNS). It could inhibit TNF-a, IL-1β, IL-6, translocation of NF-kB p65, oxidative stress, reduced the expression of caspase-3 and Bax, and increased Bcl-2 in cerebral cortex cells after cardiopulmonary resuscitation and brain trauma rats, The same effect was seen in hippocampal damage caused by paraquat poisoning, isoflurane anesthesia, and ischemia-reperfusion25,26,27,28. In this study, the anti-inflammatory and anti-apoptotic effects of ulinastatin on cerebral cells were demonstrated in Escherichia coli meningitis rats.

Protein kinase C alpha (PKCα) is a member of the PKC family of serine/threonine kinases and has been implicated in the regulation of cell survival and apoptosis in neurons.The effects of PKC activation can vary significantly depending on the specific PKC isoform involved. While isoforms such as PKCεare generally associated with anti-apoptotic effects29, other isoforms like PKCδ and PKCα can promote apoptosis30. PKCα can influence apoptotic pathways by affecting various signaling cascades. It may activate or enhance the activity of pro-apoptotic proteins while inhibiting anti-apoptotic factors, leading to cell death. PKCα has been reported to alter the balance between pro-apoptotic and anti-apoptotic members of the Bcl-2 family. For example, PKCα may increase the expression of pro-apoptotic proteins like Bax while decreasing anti-apoptotic proteins like Bcl-2. PKCα’s role in cell cycle regulation can also affect apoptosis. By disrupting cell cycle checkpoints or promoting cell cycle arrest, PKCα can contribute to apoptotic processes31.

Immunohistochemical staining immunofluorescence and western blotting results showed decreased PKCα phosphorylation (p-PKCα) and increased ZO-1 expression after ulinastatin intervention, with PKC agonists blocking ulinastatin action. Another animal experiment confirmed that ulinastatin increased ZO-1 expression in LPS -induced lung injury32. ZO-1 was initially identified as part of the tight junction in 198633, linking occludin and claudins to the actin cytoskeleton34. ZO-1 expression was significantly reduced in cerebral cells of rats with E.coli meningitis15,35, ZO-1 serine/threonine phosphorylation disrupts BBB integrity and permeability. PKC is an important signaling molecule in the regulation of life activities, which is divided into multiple isoforms and plays an important role in inflammation. PKC activation is mainly divided into two parts, one is the release of pseudosubstrate from the substrate binding lumen, and the other is the phosphorylation of the catalytic region. PKC is a classical serine/threonine kinase, and ulinastatin is a kuntiz domain that inhibits phosphorylation of serine/threonine. PKCα could be activated by LPS and alter brain microvascular endothelial cells (BMECs) tight junctions, with greater activity of PKCα, the tight junctions lose their ability to form a tight barrier36. The host cell signal transduction pathways of PKC were involved in meningitis-causing E. coli invasion of HBMECs37. our Immunohistochemical and western results suggested that ulinastatin inhibited the phosphorylation of PKCα and reduced the degradation of ZO-1 in E.coli meningitis rats, what’s more, the affection for ulinastatin is blocked by PKC agonists. In this study, ulinastatin acted as a PKC-like inhibitor, inhibiting PKCα activity and likely inhibiting E.coli invasion into the HMEC and protecting ZO-1 expression, thus protecting the blood–brain barrier and having neuroprotective effects on the brain.

This study provides valuable insights into the effects of ulinastatin, but several limitations should be acknowledged. Firstly, the sample size used in our study was relatively small, which may limit the generalizability of our results. Future studies with larger cohorts are needed to validate our findings. Secondly, the study primarily assessed the short-term effects of ulinastatin. Long-term effects and potential side effects were not evaluated, which are important for understanding the overall efficacy and safety of the treatment.

Conclusion

In conclusion, our findings suggested that ulinastatin achieved neuroprotection by inhibiting PKCα phosphorylation and increasing ZO-1 expression, reducing brain cell apoptosis in a rat model of E. coli meningitis. Therefore, Ulinastatin can be considered as a strategy for therapeutic intervention in bacterial meningitis.