A total of one hundred rhizosphere clay soil samples were collected from Sharkia Governorate, Egypt. The samples were spread onto starch-nitrate agar medium supplemented with 50 μg/ml of cyclohexamide and nystatin (Oxoid, Cambridge, UK) and incubated at 37°C for 5 days (Bae et al., 1972). After incubation, a grey colony that secretes red pigment was selected and purified by repeated streaking onto starch-nitrate agar medium. The recovered isolates were identified using the International Streptomyces Project’s descriptions (ISP) of mycelium shape, color, substrate mycelium, melanin production, and the soluble pigment (Shirling and Gottlieb, 1966). The analysis of the sugar components in the whole-cell hydrolysate and the isomer of diaminopimelic acid (DAP) within the cell membrane was performed as previously described (Marmur, 1961). The micromorphological characteristics of the isolates grown on starch-nitrate agar at 28°C for 14 days were observed using a bright-field light microscope (Nikon, China) and a scanning electron microscope (Joel, JSM-6360LA, Japan).

The active isolates were confirmed molecularly by PCR amplification of 16S rRNA gene using forward primer (5′-AGAGTTTGATCMTGGCTCAG-3′) and reverse primer (5′-TACGGYTACCTTGTTACGACTT-3′) (Lagacé et al., 2004). The amplicons were size-confirmed by electrophoresis using 1.2% agarose gel (Applichem, Germany, GmbH) stained with 0.5 μg/ml ethidium bromide (Sigma-Aldrich, MO, USA), and the identities were further verified by DNA sequencing in the forward and reverse directions on an Applied Biosystems 3130 automated DNA sequencer (ABI, 3130, USA) using a ready-reaction Bigdye Terminator V3.1 cycle sequencing kit (Perkin-Elmer/Applied Biosystems, Foster City, CA; Cat. No. 4336817). Using the Basic Local Alignment Search Tool (BLAST) on the National Center for Biotechnology Information (NCBI) website, the DNA sequences were compared to the GenBank sequences (www.ncbi.nlm.nih.gov). The MEGA11 (www.megasoftware.net) was used to analyze nucleotide sequences.

A total of 143 non-duplicate isolates including 14 Staphylococcus aureus, 8 Listeria monocytogenes, Bacillus cereus ATCC 36621, 12 Streptococcus equi, 4 Streptococcus pyogenes, 7 Streptococcus agalactiae, 17 Escherichia coli, 2 Pseudomonas aeruginosa, 13 Salmonella enterica, 11 Klebsiella pneumoniae, 17 Aeromonas hydrophila, and 1 Flavobacterium columnare, 19 Candida albicans, 3 Cryptococcus neoformans, 3 C. gattii, 3 Aspergillus flavus, 7 A. niger, and 1 A. fumigatus were involved in this study. The source of each isolate is listed in Table 1S . All isolates were identified using phenotypic and molecular methods using genus- and species-specific primers ( Table 2S ) and kept at −20°C for subsequent use in brain heart infusion broth or Sabouraud dextrose broth (Oxoid, USA) containing 20% glycerol. Serotyping of Salmonella isolates was carried out using commercially available antisera (Denka Seiken Co., Ltd., United Kingdom) in accordance with the antigenic profile (Kauffmann, 1957).

The susceptibility of 143 isolates to the commonly used antimicrobials was examined using the disc diffusion method. Inhibition zone diameters for each antimicrobial determining a resistant, intermediate, or sensitive result were interpreted following the Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI, 2020). To confirm disc diffusion results, minimum inhibitory concentrations (MICs) were determined for colistin, ciprofloxacin, tigycycline, and vancomycin against Gram-negative bacteria, Salmonella isolates, and S. aureus, respectively, using the broth microdilution method (CLSI, 2020). CLSI recommended MICs breakpoint for ciprofloxacin, tigycycline, colistin, and vancomycin were used (≥ 1 µg/ml, >2 µg/ml, ≥ 4 µg/ml, and > 8 µg/ml, respectively) (CLSI, 2020).

Streptomyces species isolates were evaluated for their antimicrobial activity toward the tested bacteria and fungi isolates using the agar plug method (Ekundayo et al., 2014). Briefly, an inoculum (1.5× 10⁸ CFU/mL) of fresh bacterial or fungal culture was inoculated on the surface of Muller-Hinton agar (MHA) plates or Sabouraud dextrose agar (Thermo Fisher Scientific Oxoid Ltd., Basingstoke, Hampshire, United Kingdom), and then Streptomyces colonies that had been cut with a sterile cork borer (8 mm) were deposited. The inoculated plate was incubated at 37°C for 24 h for bacteria and 28°C for 48 h for fungi. The results were recorded as “very strong” (zone diameter ≥ 31 mm), “strong” (21-30 mm), “moderate” (11-20 mm) or “weak” (10 mm) (Sahin and Uğur, 2003).

A slant culture of each Streptomyces species was inoculated in 100 mL ISP-2 medium in a 200-mL Erlenmeyer flask and incubated at 30°C for 5 days on an orbital shaker at 200 rpm. The broth culture (100 mL) was centrifuged at 8000 rpm for 15 min, shaken vigorously on a flask shaker, and then filtered using filter paper (Whatman No. 1). Then ethyl acetate fractions of S. misakiensis metabolites and diethyl ether for S. coeruleorubidus metabolites were concentrated using a rotary vacuum evaporator (Micro Technologies, Myanmar, United States) at 40°C. The crude extracts were resuspended in 1 mL of 96% ethanol for washing and kept at −20°C (Mohamed et al., 2017).

The antimicrobial activity of extracellular crude metabolites was determined by an agar-well diffusion test (Chaudhary et al., 2013). The tested organism was adjusted to 1.5× 10⁸ CFU/mL and then inoculated on MHA plates. Wells were made using a sterile cork borer number 4 and 100 µL of each metabolite was added. Plates were incubated at 37°C for 24 h (for bacteria) or 28°C for 48 h (for fungi). The diameters of the inhibition zone were measured and interpreted in accordance with Sahin and Uğur (2003).

Thin-Layer Chromatography (TLC) (Sigma-Aldrish, Germany) was used to evaluate the extract initially on silica gel paper chromatography. To choose the solvent system capable of demonstrating greater resolution, TLC was conducted using various solvent systems with various polarities, such as chloroform: ethyl acetate (3:9) and toluene: ethyl acetate: formic acid (7:3:0.2) for both Streptomyces species metabolites. The aforementioned Streptomyces metabolite extracts were administered via capillary tubes to pre-coated TLC plates and then developed in a TLC chamber using the appropriate mobile phase. A 10 µL aliquot of the extract solutions was applied to the TLC paper. The produced TLC plates were air dried before being inspected in the UV TLC viewer under UV at both 254 nm and 366 nm. The rate of flow (Rf) values of the observed spots were calculated (Gaurav et al., 2020). Each band was scraped off individually, extracted with methanol, and put into a distinct vial. Then, an agar-well diffusion test was used to check each band’s antibacterial and antifungal activities. The preparative high-performance liquid chromatography (HPLC) (Good Science Instrument Technology Co., Ltd., Tianjin, China) was used to identify the active eluent compounds from TLC plates (Maleki and Mashinchian, 2011).

Gas chromatography-mass spectrometry (GC-MS) was used to examine the metabolites. It was done with a thermal mass spectrometer detector (ISQ Single Quadrupole Mass Spectrometer) and a TRACE GC Ultra Gas Chromatograph (THERMO Scientific Corp., USA). A TR-5 MS column (30 m x 0.32 mm i.d., 0.25 μm film thickness) was installed in the GC-MS system. The following temperature program was used for the analyses, which used helium as the carrier gas at a flow rate of 1.0 mL/min and a split ratio of 1:10: 60°C for 1 min, followed by a 4.0°C/min increase to 240°C and a 1 min hold. At 210°C, the injector and detector were maintained. One μL of the diluted samples (1:10 hexane, v/v) was injected. Using a spectral range of 40–450 m/z and electron ionization at 70 eV, mass spectra were produced. AMDIS software (www.amdis.net) was used to identify the chemical components of the metabolite, which were then determined by their retention indices (relative to n-alkanes C8-C22), mass spectra matching to standards (when available), Wiley spectral library collection, and National Institute of Standards and Technology (NIST) library database (Managamuri et al., 2017). The obtained data were verified using IR and NMR.

An IR spectrophotometer (ThermoFisher Nicolete IR IS10-USA) was used to scan the compounds’ infrared spectra at a range of 400 and 4000 cm^-1 using ethyl acetate solution for S. misakiensis and diethyl ether for S. coeruleorubidus metabolites. Plots of the spectra’s intensity and wave number were made. Maximum and minimum resolutions, as well as the number of peaks, were measured in this spectrum range (Maleki and Mashinchian, 2011). Nuclear magnetic resonance (NMR) spectroscopy (Thermo Scientific Multiskan Sky High Microplate Spectrophotometer, Germany) is an analytical technique for obtaining detailed structural and quantitative information on metabolites. The data obtained were compared with those of the similar compounds produced by S. misakiensis and S. coeruleorubidus (Motohashi et al., 2008).

The broth microdilution method was performed using 96-well polystyrene microtitre plates (Costar, Corning Inc., USA) for the detection of MIC values of Streptomyces species metabolites against the tested bacterial and fungal isolates (Al-dhabi et al., 2020). Vacum dried ethyl acetate or diethyl ether metabolites of Streptomyces species were dissolved in tween 20 (1 gm/1 ml) then diluted (concentration range: 0.125 µg/ml to 512 µg/ml) in Mueller-Hinton broth in the case of bacterial isolates or in RPMI 1640 medium (Thermo Scientific™ Oxoid, Ltd., Basingstoke, Hampshire, United Kingdom) for testing fungal isolates. Subsequently, each well was inoculated with 100 µL of fresh bacterial culture suspension (5 × 10⁵) or fungal suspension (5 × 10⁶ CFU/ml) and the 96-well microtitre plate was incubated for 24–48 h at 37°C. Positive and negative controls were included. The lowest concentration that inhibits bacterial or fungal growth was established as the MIC.

The effective secondary metabolite identified in vitro was tested in vivo in S. aureus and K. pneumoniae infection mouse models. A total of 112 five-week-old Albino mice (20–25g) were used for the septicemia and pneumonia infection models. They were kept in animal rooms at 25°C and allowed to acclimatize for 2 weeks prior to the start of the experiment to exclude any infection. The protocol of this experiment was approved by the Institutional Animal Care and Use Committee at Ain-Shams University (approval number ASU-SCI/MICR/2023/1/4).

For calculation of the 50% effective dose (ED50), five doses of the metabolite extract (100 mg/kg of body weight used as the highest dose) were created by using sequential 1.414-fold to 1.732-fold dilutions. Immediately after infection with S. aureus or K. pneumoniae, metabolite extract was injected into the tail veins of mice in the two groups. The survival rate on day seven following infection was used to calculate ED50s and 95% confidence intervals (Otani et al., 2003).

Mice were divided into seven groups (7 animals/group). Animals in groups G1, G2, and G3 were injected intraperitoneally (I.P.) with 0.2 ml of S. aureus suspension (1.5×10⁸) in phosphate buffered saline (PBS) (Otani et al., 2003). The first group (G1) was the control positive, non-treated group; G2 was infected and treated with 20 mg/kg of body weight gentamicin (GEN, Fulford, India, Ltd.) via the tail vein for seven days (Fierer et al., 1990); and G3 received 10 mg/kg of body weight metabolite extract dissolved in tween-20 (Sigma Aldrich, St. Louis, MO, USA) through the tail vein immediately after infection for seven days.

G4, G5, G6, and G7 were the negative control groups that received 0.2 ml of saline, tween-20 intraperitoneally, 10 mg/kg metabolite extract, and 20 mg/kg GEN, respectively for seven days. At day 7 postinfection, mice were sacrificed, and the liver and spleen were harvested for analysis of the bacterial burden. Aliquots of each organ homogenate in PBS were serially diluted, plated onto Mannitol Salt Agar and Baird-Parker Agar media (Oxoid, Cambridge, UK), incubated overnight at 37°C, and observed for viable colonies.

The experiment protocol of Rodgers et al. (2019) was followed, which involved giving mice 50 µl of K. pneumoniae culture suspension in PBS (1.5×10⁸) intranasally, resulting in dissemination 24 hours after infection. The first group (G1) was the control positive group, G2 received 7.5 mg/kg GEN I.M. via the hind thigh muscle, and G3 received metabolite extract twice on the second day postinfection and then once a day for the next two days. The control negative groups (G4-G7) received 0.2 ml saline, tween-20 I.P., metabolite extract (7.5 mg/kg, I.M.), and GEN [7.5 mg/kg, I.M. (Rodgers et al., 2019)]. The antibacterial efficacy of metabolite extract in comparison with GEN for treating pneumonia was evaluated after three days postinfection based on the clinical signs and mortality rate that were recorded daily for three days. The total bacterial count of K. pneumoniae was estimated by a ten-fold serial dilution of the lung samples of mice that received GEN, metabolite extract and control groups. The appropriate dilutions were inoculated on MacConkey’s and eosin methylene blue (EMB) (Oxoid, Cambridge, UK) agar plates and incubated for 24 hours at 37°C.

The survival rate (SR) of mice in various groups was determined using the following formula: SR = Total number of mice/Total number of mice mortalities ×100 (Collins and Kays, 2014).

Blood samples were obtained from mouse groups in the septicemia model at day 7 postinfection and from the pneumonia model at day 3 postinfection for the determination of clinical chemistry parameters. Firstly, the mice were anaesthetized with 5 mg xylazine and 100 mg ketamine (Sigma Aldrich, St. Louis, MO, USA) (I.M.) (Parasuraman et al., 2010). Approximately 2 ml of blood were drawn from each mouse via the caudal vertebral vein with a 22 mm needle and immediately transferred into sterile tubes (Johnson & Johnson, Ramsey, MN, USA), which were allowed to coagulate and then centrifuged at 3500 rpm for 5 min to obtain serum. The serum total protein (TP), aspartate aminotransferase (AST), Alanine transaminase (ALT), urea, and creatinine levels in different animal groups were determined using Spinreact kits (Esteve De Bas, Girona, Spain) at a wavelength of 540 nm, in accordance with the methods of Kamimoto et al. (1985). Additionally, the consequences of renal damage such as urea and creatinine were measured using Spinreact kits (Esteve De Bas, Girona, Spain) in accordance with the methods established by Fossati and Prencipe (2010) and Murray (1984), respectively, by an automatic biochemical analyzer (Hitachi 902, Roche Diagnostics).

The collected specimens from the liver and spleen from different groups in the septicemia model and the lungs from the pneumonia model were fixed in 10% neutral buffered formalin, dehydrated in ascending grades of alcohol, cleared in xylene, and embedded in melted paraffin wax. Paraffin 5-µm sections were obtained using a microtome (Thermo Scientific, Massachusetts, USA) and stained with hematoxylin and eosin (Suvarna et al., 2018).

All the experiments were done in triplicate, and the results were displayed as means ± standard error (SE). A Shapiro-Wilk test was conducted in order to check for normality as described by Razali and Wah (2011). The data were analyzed using one-way analysis of variance (SAS Institute Inc, 2012). The differences between the survival rates of the studied groups were examined using the Chi-Square test (χ2) (SAS Institute Inc, 2012). Statistical significance was set at a P-value less than 0.05.

Among the 58 Streptomyces isolates screened, two isolates displayed antimicrobial activity. Streptomyces species isolates grew well on some media but only moderately or barely on others. Streptomyces misakiensis grew well and produced gray-white aerial mycelium on starch-nitrate agar medium, but the substrate mycelium was red and did not produce soluble pigment. S. misakiensis grew moderately on yeast malt agar medium (ISP-2) and produced white-gray aerial mycelium but no substrate-soluble pigment. On glycerol asparagine agar medium (ISP-5), S. misakiensis grew slowly with grey aerial mycelium. On tyrosine agar medium (ISP-7), S. misakiensis gave weak growth, yellowish-gray, white aerial mycelium, but the substrate mycelium was dark yellow without soluble pigments. On medium (ISP-6), the isolate gave moderate growth and pale gray aerial mycelium, but the substrate mycelium was yellowish without soluble pigment. Melanoid pigmentation and soluble pigmentation were not seen. The colour of S. coeruleorubidus conidia was grey on all ISP media except ISP-6, which was nil. High sporulation rates were observed on ISP-2 medium; good sporulation was found on ISP-3 and 4; and low sporulation was found on ISP-7.

The ability of S. misakiensis isolate to utilize various carbon compounds as the source of energy was indicated by heavy growth on ISP media supplemented separately by the following carbon sources: glucose, arabinose, xylose, inositol, mannitol, fructose, rhamnose, sucrose, and raffinose. S. coeruleorubidus has the ability to grow on ISP media supplemented with all tested carbon sources. A 14-day-old S. misakiensis culture on starch-nitrate agar medium was examined microscopically to exhibit short filamentous mycelia with few straight and smooth surface spore chains ( Figure S1A ). S. coeruleorubidus spore chains were spiral with a hook end and a spiny surface ( Figure S1B ).

DNA sequencing of 1485-bp amplicons of the 16S rDNA gene ( Figure S2 ) confirmed the identification of S. misakiensis and S. coeruleorubidus (GenBank accession numbers OP168477 and OP168352, respectively).

Preliminary antimicrobial activity screening of both Streptomyces species revealed that S. misakiensis has strong antimicrobial activity against Gram-positive bacteria (S. aureus, L. monocytogenes, B. cereus, S. equi, S. pyogenes, and S. agalactiae) as well as Gram-negative bacteria (E. coli, Salmonella species, K. pneumoniae, F. columnare, A. hydrophila, and P. aeruginosa). However, S. coeruleorubidus showed strong to moderate antimicrobial activity on S. agalactiae, S. pyogenes, L. monocytogenes, B. cereus, A. hydrophila, E. coli, K. pneumoniae, and Salmonella enterica, but didn’t inhibit P. aeruginosa and F. columnare. Furthermore, S. misakiensis has strong antifungal activity against A. flavus, A. niger, C. neoformans, C. gattii, and C. albicans, whereas S. coeruleorubidus has moderate to strong antifungal activity ( Table 1 and Table 1S ). The secondary metabolites of both species were purified and identified for further application in an in vivo model.

TLC analysis of Streptomyces metabolite extract revealed six fractions. The two fractions, R2 and R3, showed antifungal and antibacterial activities, respectively, against the tested species. The Rf value of R2 was 4.5 cm and 3.5 cm for R3. The TLC results were confirmed by HPLC preparative, which gives six peaks as shown in Figure S3 : 18.9, 19.409, 20.228, 21.615, 24.727, 29.392, 32.16, 34.7, and 35.2.

The GC-MS analysis of the purified compound from TLC was performed. The chemicals were identified based on their peak area, molecular weight, and molecular formula. The amount of substance in the active band is exactly proportional to this area. Ursolic acid methyl ester (Urs-12-en-28-oic acid, 3-hydroxy-, methyl ester, (3á)) is a novel antibacterial metabolite identified from S. misakiensis metabolites extrac, whereas tetradecamethylcycloheptasiloxane is the identified antifungal compound, as shown in Figures S4 , S5 .

The IR spectrum of the ursolic acid methyl revealed functional group spectra at: 32982, 1748, 1377, 1243, 1051, 929, 847, 620, and 456.18 cm^-1 for ( OH) stretching alcohol, ( C-H aliphatic), ( C═ O), (ester) and ( C═ C), respectively ( Figure S6 ). As shown in Figure S7 , the 1H NMR spectrum was at 3.502, indicating that this antibiotic is mostly ursolic acid methyl ester (Urs-12-en-28-oic acid, 3-hydroxy-, methyl ester, (3á)). As shown in Figure S8 , the IR spectrum of tetradecamethylcycloheptasiloxane displayed functional group spectra at 2981, 2865, 1056, 1033, 1010, and 520 cm^-1. The NMR spectrum was at 80.02, as revealed in Figure S9 , indicating that this antibiotic is mostly tetradecamethylcycloheptasiloxane.

S. coeruleorubidus diethyl ether extract was separated on TLC plates into three fractions (A, B, and C). The one active fraction Rf value for the strain was 2.0 cm, which has antibacterial and antifungal activity against the tested bacteria and fungi. HPLC analysis gives five peaks at 19.409, 20.228, 21.615, 24.727, and 29.392 ( Figure S10 ).

The metabolites of S. coeruleorubidus’ diethyl ether extract were analysed using GC-MS, and the results revealed that the antibacterial and antifungal compound thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester, predominated ( Figure S11 ). Figures S12 , S13 depict the IR and NMR spectra of thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester.

Antibacterial and antifungal activities of the bioactive compounds in terms of MIC are shown in Table 1 . Ursolic acid methyl ester exhibited antibacterial activity against a variety of species of Gram-positive and Gram-negative bacteria with MIC values ranging from 0.125 to 2 μg/ml, whereas thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester showed antimicrobial activity against E. coli, K. pneumoniae, A. hydrophila, and S. pyogenes (MIC 0.5 - 4 μg/ml). Some strains of Salmonella enterica serovars and L. monocytogens were inhibited with thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester (MIC values of 1-8 μg/ml and 4-8 μg/ml, respectively). S. aureus, S. equi, and P. aeruginosa were resistant to thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester. Concerning antifungal potential of both Streptomyces species metabolites extracts, MICs ranged from 0.125 to 1 μg/ml for tetradecamethylcycloheptasiloxane and 1 to 16 μg/ml for thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester. Tetradecamethylcycloheptasiloxane was more effective on C. albicans than thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester (MIC 0.125-2 μg/ml vs 1-8 μg/ml). Moreover, tetradecamethylcycloheptasiloxane was highly effective on C. neoformans and C. gattii but thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester was not effective on both Cryptococcus species. The MIC₅₀ and MIC₉₀ of the antibacterial ursolic acid methyl ester and the antifungal tetradecamethylcycloheptasiloxane against all tested bacteria and fungi were 0.5 μg/ml and 1 μg/mL, respectively, versus thiocarbamic acid, N,N-dimethyl, S-1,3-diphenyl-2-butenyl ester against bacteria (MIC₅₀: 2 μg/ml and MIC₉₀: 4 μg/mL) and fungi (MIC₅₀: 4 μg/ml and MIC₉₀: 8 μg/mL).

As presented in Figure 1A , at day 7 postinfection, the total bacterial burden in the liver sample of infected, non-treated G1 was 9.75 log CFU/g, followed by GEN-treated G2 (7.76 log CFU/g). The lowest bacterial count recorded was 5.75 log CFU/g in ursolic acid methyl ester–treated G3.

The highest total bacterial count in the spleen sample was 9.65 log CFU/g in G1, followed by G2 (7.54 log CFU/g). Ursolic acid methyl ester-treated G3 has the lowest bacterial count (5.67 log CFU/g). There were significant differences between G1 and G2, G2 and G3, and between G1and G3 (P < 0.05) ( Figure 1A ).

There was a significant decrease in log CFU in liver tissue of GEN-treated mice compared to the infected, non-treated G1 (1.99 log CFU; P < 0.05). Similarly, there was a significant decrease in log CFU in the ursolic acid methyl ester–treated G3 compared to the control group (4.00 log CFU; P < 0.05). However, the group of mice treated with GEN was significantly higher in log CFU (2.01 log CFU; P < 0.05) than their counterparts treated with ursolic acid methyl ester (G3).

The GEN treated G2 had significantly lower log CFU (2.11 CFU; P < 0.05) in the spleen than the untreated group (G1). Furthermore, those treated with ursolic acid methyl ester had 3.98 fewer log CFU than the control group (P < 0.05). In comparison to the GEN-treated G2, the ursolic acid methyl ester-treated G3 reduced log CFU by 1.87 (P < 0.05; Figure 1A ).

At day 3 postinfection, the microbial load in the lung sample in the pneumonia model, as shown in Figure 1B , was 9.88 CFU/ml in G1, 7.86 log CFU/ml in G2, and 7.68 Log CFU/ml in G3. There was a significant difference (P < 0.05) between G1, G2, and between G2, G3 and G1, G3.

In the septicaemia model, there were significant differences (P = 0.0324) between the survival rate in the examined groups, being 100% in ursolic acid methyl ester-treated G3, 80% in GEN-treated G2, and 20% in the infected, non-treated group G1. Regarding the pneumonia model, survival rates varied between 15% in G1, 70% in G2, and 100% in G3 (P = 0.0168).

ALT and AST levels were high in both infected, non-treated G1 (52.87 ± 0.975; 63.33 ± 0.698 mg/dL) and GEN-treated G2 (84.37 ± 0.369; 92.17± 1.33 mg/dL) in the septicaemia model. Moreover, high liver enzyme levels were found in G1 (42.49 ± 0.606; 58.27 ± 0.433 mg/dL) and G2 (57.43 ± 1.131; 65.23 ± 1.068 mg/dL) in the pneumonia model. Ursolic acid methyl ester-treated G3 showed normal liver enzyme levels compared to GEN-treated G2 that showed high levels in septicaemia and pneumonia models, as listed in Tables 3S , 4S . In the septicaemia and pneumonia models, all control negative groups, including tween-20 (G7), saline (G6), and ursolic acid methyl ester (G4), showed normal liver enzyme levels.

The total protein and albumin levels were low in G1 and GEN-treated G2 compared with the control negative groups. The total bilirubin and direct bilirubin were high in G1 and GEN-treated G2, followed by GEN-control negative G5. There were significant (P < 0.05) differences between all groups and the ursolic acid methyl ester-control negative G4 ( Tables 3S , 4S ).

In the septicemia model, the highest levels in urea and creatinine were found in GEN-treated G2 (61.90 ± 0.606; 2.30 ± 0.254 mg/dL), infected, non-treated G1 (58.50 ± 0.952; 1.87 ± 0.086 mg/dL), and GEN-control negative G5 (58.44 ± 0.594; 1.90 ± 0.054 mg/dL). Ursolic acid methyl ester G3 showed normal urea and creatinine levels (40.33 ± 0.282; 1.13 ± 0.077 mg/dL) versus GEN-treated G2 that showed high levels.

Regarding animals in the pneumonia model, GEN-treated G2, GEN-control negative G5, and G1 recorded the highest urea and creatinine levels ( Tables 3S , 4S ). Although animals in ursolic acid methyl ester-treated G3 (38.87 ± 0.588; 0.80 ± 0.057 mg/dL) have normal urea and creatinine levels, GEN-treated G2 displayed higher levels (58.13 ± 1.062; 1.93 ± 0.069 mg/dL). Normal levels of both urea and creatinine were detected in the control negative groups (G4, G6, and G7) of the septicemia and pneumonia models ( Tables 3S , 4S ).

Multiple areas of coagulative necrosis surrounded by intense inflammatory cell infiltrates and focal areas of microabscesses composed mainly of neutrophils were observed in G1. Most hepatic cells revealed degenerative changes such as hydropic degenerations and steatosis. Some hepatic blood vessels exhibited the presence of thrombus, which formed from a mass of fibrin threads and leukocytes attached within the tunica intima ( Figures 2A, B ). GEN-treated G2 showed restoration of most hepatic parenchyma with replacement of focal areas by leukocytic infiltrates ( Figure 2C ), while the liver of ursolic acid methyl ester-treated G3 showed normal hepatic structures with the presence of a few perivascular inflammatory cell infiltrations ( Figure 2D ). The liver of control negative groups (G4, G5, and G6) showed normal histomorphological structures of hepatic cords, central veins, and portal areas ( Figures 2E–H ).

The spleen of G1 contained fewer white pulp cells due to eosinophilia and basophilic granular necrotic materials. Some white pulp lymphoid populations have been replaced by cystic formations surrounded by necrotic debris. Most examined sections contained a high number of megakaryocytes ( Figures 3A, B ). The spleen of GEN-treated G2 showed relatively normal white pulp with a large number of megakaryocytes within the red pulp and heavy neutrophil, lymphocyte, and macrophage infiltration within dilated splenic sinusoids and blood vessels ( Figure 3C ). In ursolic acid methyl ester-treated G3, there was apparent normal white and red pulp with some mildly depleted white pulp, as evidenced by a decreased number of lymphoid elements in both the germinal and mantle zones ( Figure 3D ). Ursolic acid methyl ester-control negative G4 showed mildly activated white pulp cells and moderately infiltrated red pulp with hematopoietic elements ( Figure 3E ). Control negative groups had normal histomorphology of white pulp lymphoid zones and red pulp with splenic sinusoids ( Figures 3F–H ).

In K. pneumoniae infection G1, histopathological examination of the lung section revealed pneumonic changes that alternated with emphysematous changes. The pneumonic changes are represented by the presence of inflammatory exudates within alveoli, necrotic parts of the epithelial lining of alveoli, and dilated blood vessels ( Figures 4A, B ). While GEN-treated G2 restored normal pulmonary tissue architectures, some of the sections examined revealed perivascular emphysema with inflammatory cell infiltrations ( Figure 4C ). In ursolic acid methyl ester-treated G3, however, normal bronchi and alveolar epithelium were observed, along with some dilated vasculatures ( Figure 4D ). The non-infected control groups (G4, G5, G6, and G7) showed normal pulmonary tissue, bronchioles, alveolar structures, alveolar ducts, and alveoli with prominent lymphoid follicles within the wall of the bronchioles ( Figures 4E–H ).