Introduction

Bovine respiratory disease (BRD) is widespread globally. It is estimated that the incidence of BRD in American farms is 16.2%1. BRD seriously endangers cattle health and causes economic losses in the cattle breeding industry2. The incidence of BRD in China has been increasing in recent years with the expansion of the breeding scale and density and changes in the breeding environment. Bacterial pathogen, including Pasteurella multocida, Mannheimia haemolytica, and Mycoplasma, are important factors causing BRD.

M. haemolytica, a common pathogen causing BRD, often colonizes the nasopharynx of healthy cattle but causes BRD under low immunity and stress conditions3. Mannheimiosis is mainly treated with antibiotics, but frequent antibiotic use can easily lead to multiple drug resistant strains, adversely affecting the treatment of such infections. Vaccination could be an effective means to deal with the disease. Current, M.haemolytica vaccines are mainly inactivated whole-cell vaccines that can only protect against single serotype infections. Therefore, subunit or live vector vaccines as supplements to the inactivated whole-cell vaccines have become research hotspots. Researchers have found that bovine M. haemolytica immunogenicity-related proteins, including OmpA, PlpE, LKT, LPS, etc., provide cattle with unsuitable immune protection. Therefore, exploring new M.haemolytica immune-related proteins has become an important research direction4,5,6,7,8,9,10,11,12,13,14,15.

NlpI is an outer membrane (Omp)-anchored lipoprotein originally identified in the Escherichia coli K-12 strain. It is believed to play a role in cell division by targeting peptidoglycans (PGs), endopeptidases (EPases), and the cell wall protein endopeptidase S (MepS) for proteolytic degradation16,17,18. Hamilton et al.19 reported that uropathogenic E. coli (UPEC) strains lacking the nlpI gene exhibit rough biofilm formation, whereas NlpI overexpression results in smooth biofilm formation, thereby altering the pathogenicity of the strain.

As an outer membrane protein, NlpI plays a role in bacterial division or cell wall degradation. The anticipated results of this study could lay a good foundation for developing new M. haemolytica vaccines.

Results

Bioinformatics analysis of the NlpI protein

The molecular formula of the NlpI protein is C1636H2503N417O476S8, its molecular weight is 5,040 Da, and its instability index is 33.28. The protein is stable, with a theoretical isoelectric point of 4.98. It has 40 negatively charged residues (Asp + Glu), 31 positively charged residues (Arg + Lys), and a half-life in mammalian reticulocytes in vitro of 30 h. The protein has an average hydrophilicity of -0.196, indicating that it is a soluble hydrophilic protein (Fig. 1A). Further analysis by Expasy ProtScale showed that the hydrophilic region was larger than the hydrophobic region. The results predicted that the NlpI protein has a transmembrane structure between amino acids 32 and 307 (Fig. 1B) and is a signal peptide. The conserved region of the gene encodes amino acids 1-789. The gene has six open reading frames (Fig. 1C). The secondary structure of the NlpI protein includes α-helix (42.81%), β-sheet (3.89%), β-turn (14.27%), and random coil (39.03%; Fig. 1D). The tertiary structure of the NlpI protein, mainly formed by α-helixes, was predicted by the SWISS-MODEL platform.The protein sequence has a similarity of 11.32% with the multifunctional virulence effector protein in the PDB database (Fig. 1E).

Fig. 1
figure 1

Partial bioinformatics results of the NlpI protein. A Hydrophilicity/hydrophobicity prediction results for the NlpI protein. B Prediction of the NlpI protein transmembrane structure. C ORFs analysis of the NlpI protein. D Prediction of the secondary structure of the NlpI protein. h, α-helix; e, β-sheet; t, β-turn; c, random coil. E The predicted tertiary structure of the NlpI protein.

Full size image

The Tools prediction showed that the NlpI protein contained 11 antigenic determinants (Fig. 2; Table 1). ABCPred prediction showed that the NlpI protein had 27 B-cell epitopes (Table 2).

Fig. 2
figure 2

Prediction of antigenic determinants for the NlpI protein.

Full size image
Table 1 Prediction of antigenic determinant for the NlpI protein.
Full size table
Table 2 B-cell epitope prediction for the NlpI protein.
Full size table

Construction of prokaryotic expression vector of the recombinant NlpI protein and protein induced purification

Consistent with the expectations, a target fragment of 924 bp was amplified by PCR using the DNA genome of M. haemolytica strain 95 as a template. The recombinant plasmid was identified by double enzyme digestion and sequencing. The sequencing results, compared by NCBI Blast, showed that the inserted gene was consistent with the target gene (Fig. 3A), indicating that the recombinant plasmid pET-32a-NlpI was successfully constructed. The phylogenetic tree results showed that the homology of this gene was 99% (Fig. 3B; GenBank accession, PQ509852).

After ultrasonication, the induced protein was centrifuged to separate between the precipitate and the supernatant. The protein was mainly found in the precipitate. The protein was purified by Ni-NTA and subjected to SDS-PAGE electrophoresis. The purification was good and resulted in a single prominent protein band at 33 kDa (Fig. 3C), which aligned with expectations.

Fig. 3
figure 3

Results of the double enzyme digestion, target gene homology, and protein purification. A Double digestion of M. haemolytica NlpI-32a. M: DL10, 000 DNA Marker; NlpI-32a: recombinant plasmid pET-32a-NlpI; N: negative control. B Phylogenetic tree of the nlpI gene of M.haemolytica. C Protein purification results. M: protein marker; 1: the purified protein.

Full size image

Reactivity analysis of the rNlpI protein

The purified rNlpI protein was detected by western blot. The primary antibody used was bovine serum obtained from cattle clinically infected with M. haemolytica, while the secondary antibody was Rabbit Anti-Bovine IgG-HRP. The results showed that the protein could bind to bovine M. haemolytica positive serum, and a clear protein band was observed, indicating that the protein had reactogenicity (Fig. 4).

Fig. 4
figure 4

Western blot results of the rNlpI protein.

Full size image

Determination of the NlpI gene transcription level at various growth stages of Mannheimia haemolytica in vitro

The mRNA transcription level of the nlpI gene at various growth time points was detected by RT-qPCR using the 16 S rRNA gene as an internal reference gene. The expression of the nlpI gene in the stationary and logarithmic phases was significantly higher than in the lag phase (P < 0.01). The expression in the stationary phase was the highest, followed by the logarithmic, decline, and lag phases (Fig. 5).

Fig. 5
figure 5

Transcription level of the nlpI gene of M. haemolytica strain 95 at the four growth phases. The group means were compared using one-way ANOVA with Dunnett’s multiple-comparison test. **P < 0.01.

Full size image

Retention of Mannheimia haemolytica antibody in BALB / c mice

Western blot on the serum of BALB/c mice collected at various stages showed that they continuously produced specific antibodies against rNlpI (Fig. 6).

Fig. 6
figure 6

The retention of anti-NlpI antibody of M. haemolytica in BALB/c mice (n = 3). A Mice infection and blood collection. B–E Western Blot was used to analyze the anti-NlpI protein antibody in the serum of mice at 7,14,21,28 days after infection with 95 strains of M. haemolytica. Images are created using GraphPad.Prism.9.5 (https://www.graphpad-prism.cn/) and combined using Adobe Illustrator 2020 (https://helpx.adobe.com/cn/download-install.html) software.

Full size image

Detection of cytokines after immunization with the rNlpI protein

No abnormal changes were observed in mice treated with the rNlpI immune adjuvant. Inflammatory assessment after immunization showed that the serum levels of TNF-α, IL-6, and IFN-γ in the NlpI and commercial vaccine groups were significantly higher than in the PBS group (P < 0.0001; Fig. 7).

Fig. 7
figure 7

Cytokine levels (IFN-γ, TNF-α, and IL-6) in mice serum at various time points post immunization (n = 4). A The immunization schedule for titer evaluation. B Cytokine levels (IFN-γ, TNF-α, and IL-6) in the serum at various time points after immunization. Two-way ANOVA was used for comparing the groups. ****P < 0.0001. Images are created using GraphPad.Prism.9.5 (https://www.graphpad-prism.cn/) and combined using Adobe Illustrator 2020 (https://helpx.adobe.com/cn/download-install.html) software.

Full size image

Evaluation of specific antibodies after immunization with the rNlpI protein

The serum IgG antibody titers increased continuously after immunization with 30 µg rNlpI and 10, 20, and 30 µg rNlpI + white oil adjuvant. The antibody level of mice in the 30 µg rNlpI + white oil adjuvant group was significantly higher than in the PBS group (P < 0.0001) and the other immunization groups. The antibody titer reached its maximum in all immunization groups on day 21. The titer was 1:800 in the 30 µg + white oil adjuvant group. IgG subtype detection showed that the NlpI protein mainly caused the body to produce the IgG2b and IgG3 antibody subtypes (Fig. 8).

Fig. 8
figure 8

IgG antibody generation against M. haemolytica strain 95 in reaction to rNlpI protein immunization (n = 10). A Immunization schedule for titer evaluation. B IgG titers against M. haemolytica strain 95, measured in the serum of BALB/c mice immunized with 30 µg rNlpI and 10, 20, and 30 µg rNlpI + white oil adjuvant; PBS and white oil adjuvant acted as controls. Measurements started seven days post injection. C IgG subtype titers (IgG1, IgG2a, IgG2b, and IgG3) against M. haemolytica strain 95 in immunized mice were measured after the third immunization. D IgG subtype detection over time inthe four mouse rNlpI protein immunization groups. Two-way ANOVA with Dunnett’s multiple comparison test was used to compare the groups. **** P < 0.0001. Images are created using GraphPad.Prism.9.5 (https://www.graphpad-prism.cn/) and combined using Adobe Illustrator 2020 (https://helpx.adobe.com/cn/download-install.html) software.

Full size image

The mice were injected intraperitoneally with M. haemolytica strain 95 seven days after the third immunization. The 20 and 30 µg NlpI + white oil adjuvant groups generated immune protection rates of 50% and 60%, respectively. The immune protection rates of the 10 µg NlpI + white oil adjuvant and 30 µg NlpI groups were under 30%. All mice in the PBS and white oil adjuvant groups died within three days (Fig. 9).

Fig. 9
figure 9

Survival rates of mice challenged with M. haemolytica strain 95 (n = 10).

Full size image

Prophylactic effects of immunization in a systemic bacterial infected model

We measured the bacterial load in the lungs of BABL/c mice that died 48 h after infection to evaluate their bacterial clearance ability after vaccination. The bacterial load of the immunization groups was significantly lower than that of the PBS group (P < 0.0001; Fig. 10). The bacterial load of the 30 µg + white oil adjuvant group was the lowest.

Fig. 10
figure 10

Detection of lung bacterial loads in immunized and non-immunized mice (n = 3). Mice were challenged with M. haemolytica strain 95 (3 × 108 CFU per mouse). We assessed the bacterial loads in the lungs. The groups were compared using one-way ANOVA with Dunnett’s multiple-comparison test. ****P < 0.0001.

Full size image

Discussion

Mannheimia haemolytica can cause bovine fibrinous pneumonia or pleuropneumonia, making it an important respiratory disease pathogen that seriously harms the global cattle industry. M. haemolytica is handled by effective prevention and control strategies and improved management. Vaccination use is an important infection control approach. In North America and Europe, M. haemolytica and other pathogens causing respiratory infections are often used to prepare vaccines to prevent and control BRD; however, their immunization efficacy should be improved20,21. M. haemolytica is common in cattle and sheep. Although antibiotics are often used in clinical treatments, long-term use is likely to cause bacterial resistance, negatively affecting the prevention and control of M. haemolytica infections. Therefore, various immune-related proteins and their immune preparations have become hot research topics.

In this study, NCBI Blast analysis of the M. haemolytica genome found that some unstudied lipoproteins had certain immunogenicity. These include the NlpI lipoprotein, which was annotated on the NCBI to play a role in bacterial division or cell wall degradation. Since NlpI was only studied in common bacteria such as E.coli, this study was the first to use it in an immunogenicity-related test for M. haemolytica. Software analysis predicted it to be a hydrophilic soluble protein with a transmembrane domain, 11 antigenic determinants, and 27 B-cell epitopes. Therefore, it was speculated that the protein has strong immunogenicity. This study used E. coli BL21 (DE3) to express the rNlpI protein. Immunized BALB/c mice were used to analyze the antigenic characteristics of the NlpI protein.

An in vitro culture of M. haemolytica showed that the nlpI gene was stably expressed throughout the four growth phases. The highest expression level was in the stable phase, followed in decreasing order by the logarithmic, lag, and decline phases. This indicates that the gene is stably and continuously expressed during bacterial survival, helping the body produce stable and lasting specific antibodies against the transcribed protein.

IgG is the main antibody in the serum. It is divided into the IgG1, IgG2a, IgG2b, and IgG3 subtypes. The Th2 humoral immune response plays an important role in preventing the invasion of pathogenic bacteria and is an important part of the body’s adaptive immune response. The Th1 cellular immune response enhances phagocyte-mediated anti-infection immunity, especially against intracellular pathogens, and promotes IgG production. Th1 and Th2 cells can promote the production of IgG2 a and IgG1, respectively22. –23 Th1 cells produce the cytokine TNF-α and IFN-γ, while the Th2 cells produce the cytokine IL-6. This study found that immunized mice produced high cytokine levels from the 4th day after the first immunization and continued producing them throughout the immunization cycle. This indicates that the rNlpI protein had no apparent systemic toxicity and could stimulate the body to activate humoral and cellular immunity. An IgG2a/IgG1 ratio > 1 in all the rNlpI immunized groups indicated that the body activated a Th1-biased cellular immune response from the 7th day of immunization. This response enhanced the body’s ability to fight M. haemolytica.

Ayalew et al.24showed that a truncated fragment containing the main antigenic determinant region of PlpE could induce mice to produce specific IgG antibodies, which mediated the complement bactericidal effect. However, that study did not perform further challenge protection tests. Li et al.10 used the truncated rPlpE protein to immunize mice twice. They found that although there were specific IgG antibodies in their serum, all mice died within 24 h after a challenge with the 1MLD ZD06 strain, indicating that the truncated rPlpE could not induce protective immunity in mice. Mice immunized with a full-length rPlpE protein showed that it could effectively reduce mortality when challenged with ZD06 virulent bacteria. Inoculation with rOmpA can induce a strong humoral immune response in mice and cattle, but the immune protection of the protein has not been further verified.24 Lu et al.25 truncated the expression of the LktA protein and immunized mice to show that the recombinant LktA protein had good immunogenicity to the homotype strain. In this study, mice were immunized with various doses of the rNlpI protein. The survival rates in the 20 and 30 µg rNlpI + white oil adjuvant groups were 50% and 60%, respectively, similar to the results of Ayalew et al.24 and Lu et al.25 We expressed the full-length rNlpI protein, achieving an immune protection effect close to that of the full-length rPlpE protein. After the challenge, the bacterial load in the lungs of immunized mice was 99.99% lower than in the control groups (P < 0.001) indicating that the protein protected the body well.

This study showed that rNlpI could induce high, specific IgG antibody generation in mice. Immunized mice effectively resisted infection by an M. haemolytica virulent strain. We conclude that rNlpI has good antigenicity and could be a candidate protective antigen against M. haemolytica. These results lay the foundation for the development of new vaccines.

Materials and methods

Bacterial strains, plasmids, and growth conditions

M. haemolytica strain 95 and pET-32a(+) were obtained from the Laboratory of Zoonosis Prevention and Control of Shihezi University. All bacterial strains were cultured in Luria-Bertani (LB) broth or on solid LB medium containing 1.5% agar.

Animals

All murine immunization experiments complied with ethical regulations for animal testing and research. The animal experiments were performed following the ' ARRIVE guidelines ‘. The Biology Ethics Committee of Shihezi University approved this study (ethical approval code A2023-228), which was performed following its guiding principles. Specific-pathogen-free (SPF) female (n = 60) and male (n = 30) BALB/c mice, weighing 24 ± 4 g, were purchased from Henan Skobes Biotechnology Co., Ltd. Immunization or orbital venous blood collection during the experiments were performed under pentobarbital inhalation anesthesia (1% pentobarbital, 125 µL/mouse ). Pentobarbital was used in excess during euthanasia.

Main test reagents

The product purification and plasmid extraction kits were purchased from Nanjing Nuoweizan Biotechnology Co., Ltd. ; DL10 000 DNA Marker, DL2 000 DNA Marker, T-Vector pMD19, T4 DNA ligase, and the restriction endonucleases XhoI and HindIII were purchased from Takara Bio; ProteinIso Ni-NTA Resin was purchased from Beijing All-Golden Biota; mouse tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interferon-γ (IFN-γ) pre-coated ELISA kits were purchased from Shanghai Yuanye Biotechnology Co., Ltd. ; the ELISA reagent was bought from Beijing Soleibao Technology Co., Ltd. ; HRP-conjugated goat anti-mouse IgG was purchased from Kang Weishi; Rabbit Anti-Bovine IgG-HRP was purchased from Beijing Solarbio Life Sciences Technology Co., Ltd.; the medium used in this experiment was purchased from Qingdao Hi-tech Industrial Park Haibo Biotechnology Co., Ltd.

Bioinformatics analysis of the NlpI protein

We analyzed the physicochemical properties of NlpI using Expasy ProtParam, hydrophobicity/hydrophilicity prediction and analysis using Expasy ProtScale, transmembrane structure prediction using TMHMM-2.0, signal peptide prediction using SignalP-5.0, open reading frame analysis using NCBI ORF Finder, protein secondary structure using Predict Protein, protein tertiary structure model using SWISS MODEL, antigen determinant prediction using Tools, B-cell epitope prediction using ABCpred, and phylogenetic tree construction for the nlpI gene coding nucleotide sequence using Mega 11.

Primer design and synthesis

Primers were designed using Premier 5 using the nlpI gene sequence published in NCBI GeneBank (Gene ID: 67368994), and the underlined restriction sites XhoI and HindIII. The upstream primer of nlpI qPCR was synthesized by Bioengineering (Shanghai) Co., Ltd. The primer information is shown in Table 3.

Table 3 The primer sequences used in this study.
Full size table

Construction of NlpI expression vector

The PCR reaction system volume was 50 µL. The reaction conditions were pre-denaturation at 95℃ for 5 min followed by 35 cycles of 94℃ for 30 s, 62℃ for 30 s, and 72℃ for 60 s, and an extension stage at 72℃ for 10 min. The PCR product was verified by agarose gel electrophoresis and gel recovery.

The target fragment was ligated with T-Vector pMD19 to construct the recombinant plasmid pMD19-T-NlpI, which was transfected into DH5α competent cells. The recombinant pMD19-T-NlpI plasmid and the pET-32a vector were double digested with HindIII and XhoI restriction enzymes, ligated with T4 DNA ligase, and transfected into E. coli BL21 (DE3) competent cells. Monoclonal colonies were selected for double enzyme digestion, PCR, and sequencing identification. The correct plasmid identified by sequencing was called pET-32a (+) -NlpI.

Induced expression of recombinant plasmid and protein purification

The recombinant plasmid was transfected into E. coli BL21 (DE3) competent cells to obtain the expression engineered bacteria. The positive colonies were collected into 1 mL LB liquid medium and diluted 1:100 in LB liquid medium for expansion culture [the Ampicillin (Amp) concentration was 100 µg/mL]. Isopropyl β-D-thiogalactoside (IPTG) at a final concentration of 0.5 mM was incubated at 16℃ overnight. Subsequently, it was centrifuged at 8,000 rpm for 15 min at 4℃, and the cells were resuspended in PBS, mixed with 5× SDS-PAGE loading buffer at a ratio of 4:1, boiled for 5 min, and detected by SDS-PAGE electrophoresis.

The samples were lysed in an ice water bath using a 60 W ultrasonic cell crusher for 30 min and then centrifuged at 12,000 rpm and 4℃ for 5 min to separately collect the sediment and supernatant. The precipitate was resuspended in PBS, and the supernatant was subjected to SDS-PAGE electrophoresis. The solubility of the recombinant protein was analyzed.

The bacterial liquid expressing the recombinant protein was purified by a 3 mL purification column containing His tag to obtain a single band of the target protein. The prepared recombinant protein was stored in a refrigerator at -80℃.

Reactivity analysis of the recombinant protein

The purified recombinant protein was mixed with 5× SDS-PAGE loading buffer at a ratio of 4:1, boiled at 95 °C for 5 min, and centrifuged at 3,000 rpm for 30 s. After SDS-PAGE electrophoresis, the products were transferred to a membrane, blocked, incubated with primary and then secondary antibodies, and observed using a gel imaging system. The primary antibody used was bovine serum obtained from cattle clinically infected with M. haemolytica, while the secondary antibody was Rabbit Anti-Bovine IgG-HRP.

NlpI transcriptional level determination

The bacterial solution was cultured overnight in 200 µL M. haemolytica bacterial solution and transferred into 20 mL BHI liquid medium for shaking culture. Bacterial suspension (2 mL) was taken after 2 h (lag phase), 5 h (logarithmic phase), 10 h (stable phase) and 24 h (decline phase).

The total bacterial RNA was extracted using the TRIzol method, and cDNA was obtained by reverse transcription. The mRNA transcription level of the nlpI gene was detected by fluorescence quantitative PCR using cDNA as a template. The fluorescence quantitative PCR reaction comprised 45 cycles of 94℃ for 30 s, 94℃ for 5 s, annealing for 20 s, and 72℃ for 10 s. Each run was repeated three times, using the 16 S rRNA gene as the internal reference gene. The relative transcription level of nlpI was calculated by the 2−ΔΔCt method.

Humoral immune response in balb/c mice challenged with Mannheimia haemolytica

BALB / c mice were infected with 1 × LD50 95 M. haemolytica strain 95. Blood samples were collected from three mice via orbital venous capillary sampling on days 7, 14, 21, and 28 post-infection. Serum was subsequently isolated from the plasma. Western blot was used to detect antibodies specific to the NlpI protein in the serum. Each well contained 5 µg rNlpI. Samples were blocked with 5% skim milk for 2 h, washed six times for 5 min in 1× TBST, incubated overnight at 4℃ with the primary antibody (serum of infected BALB/c mice) at a dilution ratio of 1:1,000, washed six times, 5 min each, with 1× TBST, incubated at room temperature for 1.5 h with an HRP-conjugated goat anti-mouse IgG secondary antibody at a dilution ratio of 1:3,000, and washed six times, 5 min each, with 1× TBST. ECL ultra-sensitive chemiluminescence substrate was used.

Detection of cytokines secreted by the immunized mice

BALB/c female mice were divided into four groups, three mice per group. The mice received subcutaneous injections on their backs on days 0, 7, and 14. The groups were injected with PBS, white oil adjuvant, 30 µg + NlpI, and a commercial vaccine. Blood samples were collected from the orbital vein on days 7, 14, 21, 28, 35, and 42. The serum was separated by centrifugation, and the concentrations of TNF-α, IL-6, and IFN-γ were measured using commercial ELISA kits.

Immunization with the rNlpI recombinant protein

Six groups of thirteen BALB/c mice animals each (five males, eight females) were injected subcutaneously on the back with 200 µL PBS, white oil adjuvant, 30 µg rNlpI, or 10, 20, or 30 µg rNlpI + white oil adjuvant on days 0, 7, and 14.

Detection of antibody titer by indirect ELISA

Capillary blood was collected from the orbital vein on days 7, 14, 21, 28 and 35. The serum was separated by centrifugation and used to detect the antibody titer. A 96-well plate was coated with 1 × 108 CFU/well of M. haemolytica for inactivation and incubated overnight at 4℃. Subsequently, 200 µL of 5% BSA blocking buffer was added to each well and incubated at 37℃ for 2 h. Mouse serum (100µL) was added to each well and incubated at 37℃ for 1 h. Diluted HRP-conjugated goat anti-mouse IgG, IgG1, IgG2a, IgG2b, and IgG3 secondary antibodies were added and incubated at 37℃ for 1 h. The optical density of each well was measured at 450 nm within 15 min after incubating with 100 µL two-component TMB color developing solution for 25 min.

Immune challenge test in mice

Seven days after injecting the third vaccine, the mice (n = 10) were injected intraperitoneally with 3 × 108 CFU/mice M. haemolytica strain 95, and their survival was assessed.

Determination of bacterial load in the lungs of immunized mice

Seven days after injecting the third vaccine, the mice (n = 3) were injected intraperitoneally with 3 × 108 CFU/mice M. haemolytica strain 95. The lungs of each mouse were taken 0.05 g and homogenized with 1 mL PBS. After allowing the samples to stand for 15 min, the supernatants was diluted at 1: 10,000 with PBS. Subsequently, 100 µL of the diluted solution was spread onto a solid BHI medium and incubated upside down at 37℃ in an incubator. The number of colonies was counted.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics for Windows, Version 26.0 (IBM Corp., Armonk, NY, USA). Statistical significance was, set at, P > 0.05 (ns), P < 0.05 (*), P < 0.01 (**), P < 0.001 (***) or P < 0.0001 (****). GraphPad Prism 9.5 was used for charting.