HJH-3 (Sequence: VNFKLLSHSLLVTLRSHL) was synthesized by Gill Biochemical Co., Ltd., Shanghai, China. Pexiganan (C122H210N32O22, MW 2477.17) was bought from Nanjing Peptide Biotechnology Co., Ltd., Nanjing, China. S. Pullorum (CVCC 533), S. Choleraesuis (CVCC 3776), S. Typhimurium (CVCC 541), Escherichia coli (E. coli, CVCC2059/1568), Staphylococcus aureus (S. Aureus CVCC 6538), and Pseudomonas aeruginosa (CVCC2087) were purchased from the China Institute of Veterinary Drug Control (CVCC, Beijing, China). E. coli (ATCC^® 25922™), S. Aureus (ATCC^®25923™), Candida albicans (C. Albicans (ATCC^®90029™) were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA). B. Pumilus (CMCC 63202), and Bacillus Subtilis (CMCC63501/R179) were purchased from the National Center for Medical Culture Collections (CMCC, Beijing, China). These clinical isolate strains of E. coli (A5), S. Pullorum (A2), S. Typhimurium (SA59), S. Typhimurium (SB323/217), S. Typhimurium (SA66/SH286), S. Typhimurium (SB209/SH96), S. Aureus (A3), and Proteus mirabilis (B7) were isolated from clinical samples and maintained by the laboratory of preventive veterinary medicine of Henan Institute of Science and Technology.

Ampicillin, agar powder, Mueller-Hinton broth (MHB), Luria-Bertani broth (LBB), and trypticase soy broth (TSB) medium were purchased from Beijing Solarbio Science & Technology Co., Ltd., Beijing, China. Formaldehyde, glutaraldehyde, and absolute ethanol were of analytical grade.

The chickens used in this study were obtained by hatching specific pathogen-free (SPF) chickens embryos in our laboratory. The grouping and treatment of chickens in the experiment are shown in Table 1. This work has received approval for research ethics from LLSC2022014 (Henan Institute of Science and Technology).

The physicochemical characteristics of the AMP HJH-3, including the isoelectric point (PI), liposolubility index, average hydrophobic index and instability index, were analyzed by the Protparam online tool.¹ The hydrophobicity of the AMP HJH-3 was predicted by the ProtScale online tool.² The structure of HJH-3 was predicted by the I-TASSER method.³

The antimicrobial activities of HJH-3 and Ampicillin were determined by measuring the minimum inhibitory concentrations (MICs) by the twofold broth micro-dilution method according to the CLSI guidance (Humphries et al., 2021). The MIC experiments were repeated three times.

Solutions of equal parts S. Pullorum at a concentration of 5 × 10⁵ CFU/mL in MHB and HJH-3 at different concentrations (0, 50, 100, and 200 μg/mL) were prepared and grown at 37°C. Samples (100 μL) were collected at 0, 1, 2, 4, 6, 8, and 24 h. These samples were diluted 10-fold in LBB. The diluted samples were plated onto LB agar plates and cultured at 37°C for 18 h. The Colony-forming units (CFU) were used to draw the time-kill kinetics curve (Yang et al., 2020). The experiments were repeated three times.

Blood was collected from the chick wing vein and centrifuged at 800 × g for 5 min. Then, the supernatant was discarded, and erythrocytes were isolated. The isolated erythrocytes were washed three times with normal saline (0.85%, pH = 7.4), and a suspension of erythrocytes was prepared. HJH-3 (200, 400, 600, or 800 μg/mL) was mixed with an equal volume of a 5% (V/V) erythrocyte suspension, and the mixture was incubated at 37°C for 1 h. Then, the hemolytic activity of HJH-3 was detected by measuring the optical density (OD) of erythrocytes at 414 nm (Norwitz et al., 2005; Wang et al., 2018). Erythrocytes treated with 1% Triton X-100 were used as positive controls (100% hemolysis). Normal erythrocytes without AMP treatment were used as negative controls (no hemolysis). The hemolysis rate of cells was calculated as follows: (OD peptide treatment—OD negative control)/(OD positive control–OD negative control). The experiments were repeated three times.

Chicken embryo fibroblasts (CEFs) were isolated and cultured according to methods reported in the paper (Beug and Graf, 1977). The effect of HJH-3 on the proliferation of CEFs was determined by the CCK-8 method (Yang et al., 2020). HJH-3 and pexiganan (50, 100, 200, and 400 μg/mL) were added to the wells. After incubating the plates for 24 h, 10 μL CCK-8 solutions were added, and the OD at 450 nm was read. Cells without peptides served as positive controls, and medium without cells served as negative controls. Cell viability (%) = [(OD_s-OD_b)/(OD_c-OD_b)] × 100. ODs = experimental well absorbance (absorbance of wells containing cells, media, CCK-8, and compounds to be tested); OD_b = blank well absorbance (absorbance of wells containing medium and CCK-8); OD_c = control well absorbance (absorbance of wells containing cells, medium, and CCK-8) (Yang et al., 2020). The experiments were repeated three times.

The initial concentration of S. Pullorum was 6 × 10⁹ CFU/mL. The bacteria were diluted to six gradients, 10^–1, 10^–2, 10^–3, 10^–4, 10^–5, and 10^–6. Forty-two 7-day-old chickens were randomly assigned to seven groups of six chickens each. Among them, six groups were treated with different concentrations of S. Pullorum, and 1 group treated with PBS served as the blank control. The challenge dose was 0.5 mL of S. Pullorum or sterilized PBS. The challenge was injected intraperitoneally (IP). The morbidity and mortality of chickens were observed after infection. S. Pullorum was isolated from diseased chickens, and the data from the chickens were confirmed. The Reed–Muench method was used to calculate the LD₅₀ (Haggett and Gunawardena, 1964; Thakur and Fezio, 1981). The calculation process was as follows:

Taking S. Pullorum as the research object, the clinical application effect of HJH-3 in the treatment of S. Pullorum infection was evaluated in a chicken model. Chicken grouping and infection are shown in Table 1. Sixty chickens were randomly divided into 6 groups: the S. Pullorum infection group (G-INF), the HJH-3 prophylaxis group (G-HJH-P), the HJH-3 treatment group (G-HJH-C), the Ampicillin prophylaxis group (G-Amp-P), the Ampicillin treatment group (G-Amp-C), and the negative control group (G-Con). For the prophylaxis group, HJH-3 or Ampicillin was used by IP (200 μg/mL, 0.5 mL), and then the chickens were infected with S. Pullorum. For the treatment group, the chickens were infected with S. Pullorum, and then HJH-3 or Amp was used by IP.

After initiation of the challenge test, chickens from each group were closely observed for clinical characteristics, such as mental status, coat condition, and fecal characteristics. Clinical symptoms of experimental chickens were divided into three levels after infection with bacteria: Mild symptoms: Lower food intake, fear of cold, body curled up, drooping wings, depression; Severe symptoms: row of white sticky or light yellow or light green loose stool, anus sometimes closed by hardened fecal mass and presence of mild symptoms listed above; Severest symptoms: no food intake, breathing difficulties, and presence of severe symptoms listed above.

After the death of the chickens, they were immediately collected and necropsied on an ultraclean Table 4, and the macroscopic changes in the tissues and organs of all systems of the dead chicks were observed and recorded. Meanwhile, the intestinal contents of dead chickens were aseptically taken to determine the cause of death. The chickens in which no death occurred during the test period were uniformly necropsied after the end of the test and handled according to the method described by Zhang et al. (2015).

The agar plate counting method was used to analyze the number of bacteria in the spleen and blood of chickens that died of infection during the experiment as described by Zhang et al. (2015). One hundred microliters of heart-collected blood was directly collected for smearing and counting. The statistical analysis of the spleen bacterial load was performed as follows. Spleens from dead chickens were removed under aseptic conditions (tissues such as the mesangium were removed as much as possible) and weighed. According to the weight of the spleen, a corresponding 20 × volume of sterilized PBS was added. Homogenization of the spleen was performed under sterile conditions, and the homogenization solution was diluted in multiple ratios. Finally, 100 μL extracts were taken for coating and counting. The experiments were repeated three times.

The deceased chickens were necropsied, and the internal organs were removed and processed for paraffin sectioning (the approximate procedure was as follows: fixation, dehydration, embedding, sectioning, staining, and blocking) according to a previously reported method (Xu et al., 2012). After that, paraffin sections were placed under a microscope to observe the organ pathology.

Statistical analyses were performed using GraphPad Prism 7.0 (Software). Unpaired Student’s t-test or Two-way ANOVA was utilized where appropriate. The threshold of statistical significance level P was set at *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

The amino acid sequence of HJH-3 was entered into the website. The physical and chemical properties of HJH-3 are as follows: molecular formula, C₉₆H₁₆₁N₂₇O₂₄; molecular weight, 2077.50 Da; theoretical pI, 11.00; positively charged residues, 2; and instability index of 18.42. The results indicated that HJH-3 is a stable peptide with an estimated half-life of 100 h (mammalian reticulocytes, in vitro), > 20 h (yeast, in vivo), or >10 h (E. coli, in vivo). The grand average of hydropathicity (GRAVY) of HJH-3 is 0.70. The results of the prediction of the structure of HJH-3 were shown in Figure 1. Its secondary structure is an α-helix.

The MIC values of HJH-3 against Gram-negative representative bacteria (E. coli and Salmonella) ranged from 1.5625 to 25 μg/mL. The MIC value for a Gram-positive representative organism (Staphylococcus) was greater than 25 μg/mL. HJH-3 also had biological activity against the representative fungus (C. Albicans), although the MIC was as high as 100 μg/mL. The results showed that HJH-3 has good antibiotic activity against Bacillus Pumilus and Bacillus Subtilis. However, HJH-3 had no antibiotic activity against Actinobacillus Pleuropneumoniae, Proteus Mirabilis, Pseudomonas Aeruginosa, or Enterococcus Faecalis at a concentration of 200 μg/mL (shown in Table 2).

As shown in Figure 2, HJH-3 at concentrations of 100 or 200 mg/mL completely killed S. Pullorum after incubation for 6 h. HJH-3 at a concentration of 50 mg/mL did not completely kill S. Pullorum after co-incubation for 6 h but did completely kill S. Pullorum after incubation for 12 h. Therefore, HJH-3 exhibited concentration-dependent bactericidal action against S. Pullorum.

The hemolytic activity of HJH-3 was detected by measuring the absorbance of erythrocytes at OD₄₁₄. The results showed that the hemolysis rate of CRBCs treated with HJH-3 was slightly higher than 10%, even for the 400 μg/mL treatment for 1 h, while the hemolysis rate of the commercial antibacterial peptide pexiganan was as high as 57.3% at 400 μg/mL (Figure 3). Although the two AMPs showed concentration-dependent hemolytic effects, the hemolytic effect of HJH-3 was approximately fourfold lower than that of pexiganan.

After 24 h of treatment with HJH-3, the CEF survival rate was assessed by CCK-8 assay. The survival rates of CEFs after incubation with HJH-3 at concentrations of 50, 100, 200, and 400 μg/mL were 95.56, 106.65, 108.78, and 112.34%, respectively. The survival rates of CEFs after pexiganan incubation at the same concentrations were 56.93, 43.45, 39.78, and 31.00%, respectively (Figure 4). The CEFs proliferation results showed that cell growth was promoted with increasing HJH-3 concentration and decreased with increasing pexiganan concentration.

The mortality of chickens infected with S. Pullorum over 7 days was shown in Table 3. The LD₅₀ of S. Pullorum was calculated according to the calculation process described above. The TCID₅₀/0.5 mL was 10^4.375. The mortality rate of chickens challenged with S. Pullorum (6 × 10⁹CFU/mL, diluted by 23713, 0.5 mL, IP) was 50%.

Clinical symptoms in the infected flocks of the different treatment groups were observed for an average of 12 h during the challenge test. The G-INF had the most severe clinical symptoms, including, within 12 h after infection, lowered or no food intake, fear of cold, body curled up, drooping wings, depression, and a preference to aggregate, with death occurring in one chick; on Day 4 after infection, row of white sticky or light yellow or light green loose stool and breathing difficulties; and all 10 chickens infected with S. Pullorum alone dying 6 days after infection. S. Pullorum were isolated from the intestinal contents of the dead chicken to ensure that the cause of death was indeed due to infection. The G-HJH-P exhibited no significant clinical symptoms for 3 days after challenge and exhibited mild symptoms in 2 chickens beginning on Day 4, which recovered after 2 days and returned to normal on Day 6. The protection rate was 100%. The G-HJH-C exhibited mild symptoms in 3 chickens on Day 2 post challenge, which recovered by Day 4, and 2 chickens died by Day 5. The protection rate was 80%. In the G-Amp-P, the best treatment effect was observed; no obvious clinical symptoms were observed in infected chicks, and no death occurred in chicks throughout the test period. The protection rate was 100%. In the G-Amp-C, the treatment effect was relatively good; 1 chicken died on the fifth day, and the protective effect was better than that in the AMP treatment group. The G-Con did not show clinical symptoms (shown in Table 4).

Chickens challenged with the bacteria were necropsied, and the most severely diseased internal organs were selected for macroscopic analysis of apparent lesions (liver, spleen, intestine, etc.). As shown in Figure 5, lesions in the intestine, spleen, and liver of chicks infected with S. Pullorum were prominent. The liver lesions were severe and dark red, and on the cut surface, blood sample foamy exudation was seen. The spleen was grossly enlarged, with a tense tunica and blunt edges. The intestinal lesions were mainly concentrated in the small bowel segments. The duodenum and cecum also had varying degrees of congestion, and the intestinal segments were engorged with ground contents and had a thin and viscous bowel wall. Visceral lesions were milder in the prevention group, and the G-INF exhibited the mildest lesions. Based on the degree of celiac lesions in the dead chickens, the G-HJH-C had a higher degree of lesions than the G-Amp-C.

The total bacterial load in the spleen of G-INF was approximately 1.65 × 10⁸ CFU/g, and the bacterial load in the G-HJH-C was approximately 3.4 × 10⁶ CFU/g, a more than 48-fold decrease. The bacterial loads in the spleens of the chicks in the G-HJH-P and the G-Amp-C were almost the same, approximately 2.0 × 10⁵CFU/g, which was significantly lower than that in the infection alone group. The lowest bacterial load in the spleen of the chicks was observed in the G-Amp-P and was approximately 1.4 × 10⁴ CFU/g (Figure 6A). The bacterial loads in the blood and spleen were largely consistent, but the number of bacteria in the blood was an order of magnitude lower than that in the spleen (Figure 6B).

The paraffin sections in different treatment groups were stained with HE. As shown in Figure 7, the liver lobules of the normal and protected chicks were intact and obvious, the sinusoidal space was obvious, there were detectable lymph nodes and central veins, and there were a few erythrocytes in the liver nucleus (Figures 7A, D). The hepatocytes of chickens infected with S. Pullorum were slightly atrophic and smaller, with fatty degeneration and vacuoles appearing, and the number of metachromatic granules increased (Figure 7G).

The normal and protected chicks had a thinner lymphatic sheath around the artery, fewer lymphocytes, an orderly arrangement of lymph nodes, and only a small amount of blood in the spleen sinus (Figures 7B, E). In chicks infected with S. Pullorum, the lymphatic sheath around the splenic artery was thickened, the number of lymphocytes was increased, and the number of lymphocytes in the glandular bodies was increased (Figure 7H).

The duodenal glands of normal and protected chicks were intact, the intestinal villi were intact, the striated edges were neat, and there were a few goblet cells in the crypt epithelium (Figures 7C, F). The capillaries of the duodenal lamina propria of chicks infected with S. Pullorum were hyperemic and swollen, resulting in congestion. The tops of villi were ruptured, and a large number of villi had sloughed off, which led to the exposure of the lamina propria. The intestinal villi were exfoliated in the intestinal cavity, and the number of goblet cells in the crypt epithelium was increased significantly (Figure 7I).