The blood samples (n=400) used in this study were initially collected for diagnostic purposes to facilitate patient care, not for research. These samples were not specifically obtained for the study. However, informed consent was obtained from all the participants to carry out this study and to maintain patient confidentiality, all clinical and laboratory data were anonymized. This study adhered to ethical standards approved by the Port-Said University Ethical Approval Committee, under approval number REC.PHARM.PSU/2023/21.

The bacterial strains employed in this study were isolated from positive blood cultures routinely collected from patients with bloodstream infections (BSIs) at the Department of Microbiology, Al-Demerdash Hospital, Egypt. Initially, the Gram-negative isolates were identified using standard microbiological techniques, which included Gram staining, biochemical tests, and cultural characteristics in addition to API (Analytical Profile Index) 20E (BioMérieux, Mary l’Etoile, France). Subsequent confirmation of these isolates was achieved using the VITEK-2 automated identification system (BioMérieux, 2021), which offers enhanced accuracy and efficiency in bacterial identification. The standard strain E. coli ATCC 25922 was used as a control strain. To preserve these clinical strains for future research, they were stored in glycerol stocks at -70°C, a method proven to maintain bacterial viability and genetic integrity over extended periods (Kumar et al., 2016; Harris et al., 2019a). This storage technique ensures the stability of the strains and facilitates their use in subsequent studies, contributing to reliable and reproducible research outcomes. Furthermore, Genetic detection of specific Gram-negative bacterial genes using traditional PCR was applied identifying pathogens including E. coli, P. aeruginosa, and A. baumannii without sequencing. This approach involves amplifying unique gene sequences associated with these bacteria to confirm their presence. For E. coli, the uidA gene, which encodes β-glucuronidase, is commonly targeted as it serves as a reliable marker for this pathogen (Sokurenko et al., 1998). In the case of P. aeruginosa, the pvdA gene, which is involved in pyoverdine biosynthesis, is utilized for identification due to its specificity to this bacterium (Schweizer, 2003). For A. baumannii, the blaOXA-51 gene, a prevalent carbapenemase gene, is frequently employed to detect this pathogen (Perez et al., 2007). The PCR process involves isolating bacterial DNA from samples, amplifying the target genes with specific primers, and visualizing the PCR products through gel electrophoresis. The presence of bands of the expected size on the gel indicates the presence of the target bacteria. This method offers a rapid, cost-effective, and reliable means of pathogen detection, facilitating timely diagnosis and treatment (Miller et al., 2009; Ghosh et al., 2012).

Clinical bacterial isolates were assessed for susceptibility to commonly prescribed antimicrobial agents for sepsis using the Kirby-Bauer disc diffusion method. In this procedure, antimicrobial discs were applied to Mueller-Hinton agar plates inoculated with bacterial isolates. The antimicrobial agents tested included: β-lactam agents with β-lactamase inhibitors such as Ampicillin/Sulbactam (SAM 10/10 µg) and Piperacillin/Tazobactam (TZP 100/10 µg); cephalosporins including Cefepime (FEP 30 µg), Cefotaxime (CTX 30 µg), Ceftazidime (CAZ 30 µg), and Ceftriaxone (CRO 30 µg); a cephalosporin and β-lactamase inhibitor combination, Ceftazidime/Avibactam (CAZ/AVI 30/20 µg); fluoroquinolones like Ciprofloxacin (CIP 5 µg); carbapenems including Ertapenem (ERT 10 µg), Imipenem (IPM 10 µg), and Meropenem (MEM 10 µg); aminoglycosides such as Amikacin (AK 30 µg); and Aztreonam (ATM 30 µg); and Colistin (CT 10 µg). Susceptibility and resistance were interpreted according to Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI, 2023; EUCAST, 2024). To confirm the results, additional antimicrobial susceptibility testing was performed to determine the minimum inhibitory concentrations (MICs) using the VITEK-2 automated system. Isolates were classified as multidrug-resistant (MDR) if they exhibited resistance to at least three different classes of antibiotics (Magiorakos et al., 2012).

In this study, several resistance genes were analyzed to evaluate their role in antibiotic resistance. bla_CTX-M (Extended-Spectrum β-Lactamase), responsible for hydrolyzing various β-lactam antibiotics, including penicillins and cephalosporins (Pitout and Laupland, 2008). bla_OXA-48 encodes a carbapenemase that confers resistance to carbapenems, crucial in treating multidrug-resistant infections (Poirel et al., 2012). bla_NDM (New Delhi Metallo-β-Lactamase) imparts resistance to nearly all β-lactams, including carbapenems, complicating treatment options (Yong et al.., 2019a). bla_SHV is associated with ESBL production, leading to resistance against penicillins and cephalosporins (Paterson and Bonomo, 2005). bla_TEM is a widely known β-lactamase gene contributing to resistance primarily to penicillins and some cephalosporins (Datta and Hughes, 1983). These genes collectively highlight the critical challenge of antibiotic resistance and underscore the need for continued surveillance and novel treatment strategies.

For the PCR detection of resistance genes in bacterial isolates, total DNA was extracted from each isolate using the JET Genomic DNA Purification Kit (Thermo Scientific, USA) according to the manufacturer’s instructions. Following the extraction of DNA, the PCR amplification was conducted using primers and cycling conditions as specified in the protocols outlined in previous literature ( Table 1 ). This approach adhered to established methods to ensure accuracy and reproducibility in the detection of specific genes. The primer sequences and cycling parameters were selected based on validated protocols from prior studies, which have been demonstrated to effectively amplify the target genes of interest ( Table 1 ). To ensure the specificity of these primers, they were analyzed using NCBI Primer-BLAST, a tool available at the NCBI website (https://www.ncbi.nlm.nih.gov/guide/data-software/). This process ensures that the primers effectively target the intended genes and minimize non-specific amplification.

Four species of desert-adapted plants, traditionally utilized for treating severe bacterial infections, were collected from their natural habitat in the Egyptian desert. These plants, detailed in Table 2 , were identified by the Herbarium of the Desert Research Center in Egypt. For each plant, 300 grams of shade-dried, coarsely powdered aerial parts were extracted using 70% ethanol through a maceration process. The evaporation of ethanol was confirmed by concentrating the resulting extracts to dryness under reduced pressure using a rotary evaporator set at 40°C. This process effectively removed the ethanol from the extracts, leaving behind the concentrated plant material, which was then stored at 4°C until further use (Ghaly et al., 2023).

The antibacterial efficacy of the plant extracts was evaluated using the agar well diffusion method. Bacterial cultures, standardized to a 0.5 McFarland turbidity equivalent, were evenly spread on sterile Mueller-Hinton agar plates to achieve uniform bacterial distribution. Plant extracts, dissolved in dimethyl sulfoxide (DMSO) (10 mg/mL), were then introduced into wells drilled into the agar at a volume of 100 µL per well. The plates were incubated at 37°C for 24 hours to allow for bacterial growth and interaction with the extracts. After incubation, the diameters of the inhibition zones surrounding each well were measured in millimeters to assess the antimicrobial activity of the extracts (Murray et al., 2016; Clinical and Laboratory Standards Institute, 2023). A positive control well containing colistin and a negative control well containing DMSO were included in the assay for comparison.

The minimum inhibitory concentrations (MICs) of plant extracts against identified Gram-negative strains were determined using the broth microdilution method. Plant extracts were initially dissolved in 10% DMSO to achieve a final concentration of 1 mg/mL in the culture broth. Serial 1:2 dilutions were then performed to obtain a range of concentrations from 1 mg/mL to 0.0156 mg/mL. Each dilution (100 µL) was added to a 96-well plate, along with negative and positive controls: culture broth with DMSO and culture broth with colistin, respectively. Each well, including the test and growth control wells, was inoculated with 5 µL of a bacterial suspension at a concentration of 10⁵ CFU/mL ( (Murray et al., 2016; Eloff, 1998).

All experiments were performed in triplicate and the microdilution trays were incubated at 37°C for 18 h. After incubation, the minimum inhibitory concentration (MIC) was determined by measuring the optical density of each well using a spectrophotometer, similar to the principles used in an ELISA assay to assess bacterial growth inhibition. MIC values were defined as the lowest concentration of each plant extract, which completely inhibited microbial growth (Valgas et al., 2007).

The checkerboard method is an established approach for determining the fractional inhibitory concentration (FIC) index, which assesses the interaction between a plant extract and a common antimicrobial drug (amikacin). This technique involves creating serial dilutions of both the plant extract and the antimicrobial drug, which are then combined in a 96-well microtiter plate, with each well containing a different concentration of the extract and drug. After adding a standardized bacterial suspension to each well, the amikacin-resistant Gram-negative bacteria, which exhibited multidrug-resistant (MDR) patterns (40 isolates with higher multiply antibiotic indices), were incubated at 37°C for 18-24 hours to facilitate bacterial growth and interaction with the compounds. The effectiveness of the plant extract and drug, both individually and in combination, is evaluated by measuring the minimum inhibitory concentration (MIC) for each in a triplicate manner. The FIC index is calculated using the formula: FIC index = (MIC of plant extract in combination/MIC of plant extract alone) + (MIC of antimicrobial drug in combination/MIC of antimicrobial drug alone). This index indicates the nature of the interaction—whether synergistic (FIC index ≤ 0.5), additive (FIC index > 0.5 but ≤ 1), or antagonistic (FIC index > 1). By revealing potential synergistic or antagonistic effects, the checkerboard method aids in optimizing combination therapies and understanding how plant extracts can enhance antimicrobial treatment (Odds, 2003; Tamma et al., 2012).

The inhibition of Cyclooxygenase-1 (COX-1) and COX-2 enzymes by the tested plant extracts was assessed using an ELISA-based method in triplicate manner. Plant extracts were dissolved in 10 µL of DMSO to create a series of stock solutions with concentrations ranging from 50 mmol/L to 500 pmol/L. Celecoxib served as the positive control. The inhibition of recombinant COX-1 and COX-2 enzymes was measured using a Cayman Human (EIA) kit (Catalog No. 560131, Cayman Chemicals Inc, MI, USA) and analyzed with a Robomik P2000 ELISA reader at 450 nm. The effectiveness of the plant extracts was quantified by determining the mean concentration required to achieve 50% enzyme inhibition (IC₅₀) ± standard deviation. Additionally, the COX-2 selectivity index (SI) was calculated as the ratio of IC₅₀(COX-1) to IC₅₀(COX-2) and compared to that of celecoxib to evaluate selectivity (Baek et al., 2021).

The antioxidant activity (AA%) of each substance was evaluated using the 2,2-diphenyl-1-picryl-hydrazyl-hydrate (DPPH) free radical assay, as described by Boly et al. (2016). In this assay, 100 μL of freshly prepared DPPH reagent (0.1% in ethanol) was mixed with 100 μL of each sample, which had been dissolved in ethanol, in a 96-well plate (n=6). The reaction mixture was incubated at room temperature for 30 minutes in the dark. The reduction in radical activity was monitored by measuring the absorbance at 540 nm using a microplate reader (FluoStar^® Omega, WWU, Münster, Germany), with ethanol serving as the blank. The percentage of scavenging activity was calculated using the formula:

Where Abs_control is the absorbance at 540 nm of 100μM DPPH solution without addition of the extract, Abs_sample is the absorbance at 540nm of 100μM DPPH with 5–100 μg/mL of sample. Trolox was used as a reference standard. The IC₅₀ value, which represents the concentration of extract required to reduce DPPH by 50%, was determined by plotting the percentage inhibition against concentration using GraphPad Prism 5^® (Chen et al., 2013).

A total of 400 blood samples were collected for blood culture analysis, resulting in the identification of 233 clinical bacterial isolates as Gram-negative pathogens, yielding a prevalence rate of 58.25%. Among these Gram-negative isolates, E. coli and K. pneumoniae were the predominant pathogens, with 113 isolates (48.5%) and 90 isolates (38.6%), respectively. The next most common pathogens were A. baumannii, represented by 16 isolates (6.9%), and P. aeruginosa, with 14 isolates (6%).

Analysis of the antimicrobial resistance patterns among these clinical isolates revealed a significant burden of multidrug resistance (MDR), as 185 isolates (79.4%) demonstrated resistance to 3 or more antimicrobial classes. Notably, P. aeruginosa had the highest MDR prevalence (92.8%). This was followed by A. baumannii, with 81.2% of isolates showing MDR patterns. Additionally, a high prevalence of MDR was observed among E. coli and K. pneumoniae, with rates of 77.9% and 78.9%, respectively. The distribution of antimicrobial resistance and resistance gene profiles among the investigated Gram-negative bacterial species: E. coli, K. pneumoniae, P. aeruginosa, and A. baumannii. were observed in Figure 1 . Our results indicated a varied distribution and heterogenous patterns of antimicrobial resistance and resistance gene profiles among different Gram-negative bacterial species. The results further revealed a heterogeneous pattern of antimicrobial resistance gene distribution, particularly in Pseudomonas aeruginosa and Acinetobacter baumannii. In contrast to E. coli and K. pneumoniae, where antimicrobial resistance and resistance genes were more consistently distributed among strains, P. aeruginosa and A. baumannii exhibit significant variability in the presence and absence of resistance genes, reflecting low clonality and high heterogeneity within these isolates.

The results for antimicrobial resistance across all Gram-negative bacteria, represented in Figure 2 , revealed substantial resistance to a wide range of antimicrobial agents. Conversely, all tested isolates demonstrated complete susceptibility to colistin, an antibiotic considered as a last-resort treatment for MDR infections. Overall, all investigated isolates displayed notably high resistance rates across various antibiotics. For amikacin, the isolates showed a moderate resistance rate of approximately 30%. However, resistance to aztreonam and ampicillin-sulbactam was significantly higher, at around 70-80%, reflecting a broad decline in the efficacy of these agents. Resistance to cefepime, cefotaxime, ceftriaxone, and ceftazidime was consistently elevated, with rates in the 70-80% range, underscoring the substantial challenges in treating septicemic patients with these commonly used beta-lactam antibiotics.

Of note, the resistance patterns among the detected Gram-negative species indicate that, for several antimicrobial, including aztreonam, ampicillin-sulbactam, cefotaxime, ceftriaxone, ceftazidime/avibactam, ciprofloxacin, ertapenem, and temocillin, the levels of resistance did not reach statistical significance, with P-values exceeding 0.05. This suggests that resistance to these antibiotics may not be markedly different across the groups analyzed. In contrast, the resistance to other antimicrobial agents showed statistically significant differences (P< 0.05), indicating that these other drugs are likely experiencing varied effectiveness due to differing resistance levels among the bacterial species tested. These findings highlight specific antibiotics that may be more or less effective in treating infections caused by these Gram-negative organisms, underscoring the importance of ongoing surveillance to guide antibiotic selection and optimize treatment strategies.

When focusing on specific pathogens, A. baumannii, E. coli, K. pneumoniae, and P. aeruginosa, the data demonstrated high levels of resistance across most tested antimicrobial agents, with A. baumannii generally showing the highest resistance rates. For amikacin, A. baumannii exhibited resistance near 60%, whereas E. coli, K. pneumoniae, and P. aeruginosa showed lower resistance levels, typically around 20-40%. Resistance to aztreonam and ampicillin-sulbactam was uniformly high across all pathogens, ranging from 70-90%, with A. baumannii and K. pneumoniae particularly high resistance levels ( Figure 2 ). The presence of resistance genes such as blaSHV, blaCTX-M, blaTEM, blaOXA-48, and blaNDM highlighted the molecular mechanisms driving resistance in these pathogens. Notably, blaNDM was highly prevalent in A. baumannii and K. pneumoniae, contributing significantly to the high levels of carbapenem resistance observed as shown in Figure 2 . Except for the presence of the blaTEM gene, all other resistance genes showed significant variation among the Gram-negative species detected (P< 0.05). These findings underscored the severe antimicrobial resistance challenges posed by these Gram-negative pathogens in septicemic patients, emphasizing the urgent need for alternative therapeutic strategies and stringent antimicrobial stewardship practices.

Ethanol extracts from medicinal plants were assessed for antimicrobial activity against MDR Gram-negative bacteria from septic patients. The agar diffusion assay revealed significant antibacterial effects, with inhibition zones ranging from 13 to 25 mm. J. candicans emerged as the most effective, showing inhibition zones of 18–25 mm, particularly strong against A. baumannii (25 mm) and E. coli (19 mm), and effective against P. aeruginosa and K. pneumoniae (18 mm each). M. ciliata also showed notable activity, especially against A. baumannii (20 mm) and E. coli (16 mm). In contrast, C. tubulosa and T. hirsuta exhibited more variable results; T. hirsuta had significant inhibition against E. coli (15 mm) and A. baumannii (14 mm), while C. tubulosa was least effective, particularly against E. coli (13 mm).

Broth microdilution testing revealed that J. candicans exhibited the highest antimicrobial potency among the tested plant extracts ( Supplementary Table S1 ). It demonstrated the lowest MIC values, with 250 µg/mL against most E. coli isolates, and 500 µg/mL against most K. pneumoniae, A. baumannii, and P. aeruginosa isolates. In contrast, C. tubulosa showed comparatively higher MIC values, reflecting a reduced effectiveness: 500 µg/mL against most E. coli and K. pneumoniae isolates, 1000 µg/mL against most P. aeruginosa, and A. baumannii isolates. M. ciliata exhibited variable efficacy, with MIC values of 500 µg/mL against most E. coli and 1000 µg/mL for most A. baumannii, K. pneumoniae, and P. aeruginosa isolates. T. hirsuta had the highest MIC values across all tested bacteria, with 1000 µg/mL against most investigated isolates. Overall, J. candicans demonstrated the strongest antimicrobial potential, followed by M. ciliata, indicating its superior effectiveness in inhibiting the growth of these Gram-negative bacteria.

The checkerboard assay, as shown in Supplementary Table S1 , assessed the combined effect of the widely prescribed antibiotic amikacin and J. candicans extract, demonstrating a synergistic interaction against all investigated amikacin-resistant, multidrug-resistant (MDR) Gram-negative bacteria (40 isolates). The combination of J. candicans with amikacin resulted in a notable reduction in the MIC values for both agents, with a FICI ≤ 0.5, indicating a significant enhancement in antimicrobial efficacy. Specifically, this combination therapy substantially lowered the MIC values for E. coli and, to a lesser extent, for K. pneumoniae, A. baumannii, and P. aeruginosa. These results suggest that J. candicans and amikacin work synergistically to inhibit bacterial growth more effectively than either agent alone. Moreover, this synergistic interaction highlights the potential of integrating J. candicans extract with amikacin as a promising therapeutic strategy to improve treatment outcomes for infections caused by resistant Gram-negative bacteria. Interestingly, over 50% of the amikacin-resistant strains were converted to sensitive phenotypes by this combination, further emphasizing the therapeutic potential of this approach.

Our results indicated that J. candicans extract exhibited the most potent inhibitory activity against both COX-1 and COX-2 enzymes, with IC₅₀ values significantly lower than those of the standard drug, Celecoxib (IC₅₀: 0.28 μg/mL for COX-1 and 1.11 μg/mL for COX-2). Specifically, J. candicans demonstrated exceptional efficacy in inhibiting these enzymes, suggesting its potential as a therapeutic agent for inflammatory conditions. In addition, the extracts of C. tubulosa and T. hirsuta also exhibited noteworthy inhibitory effects on both COX-1 and COX-2, with IC₅₀ values that closely approached those of Celecoxib, as detailed in Table 3 . Importantly, all four plant extracts showcased a COX-2 selectivity index comparable to that of Celecoxib, ranging from 0.23 to 0.28. This selectivity indicates a preference for inhibiting COX-2 over COX-1, which is advantageous for minimizing gastrointestinal side effects often associated with non-selective COX inhibitors. The ability of these extracts to selectively target COX-2 further emphasizes their potential as valuable anti-inflammatory agents. Overall, the findings suggest that these plant extracts, particularly J. candicans, not only provide strong anti-inflammatory effects but also present a promising alternative to conventional NSAIDs. Further investigation into their mechanisms of action and clinical applications is warranted to fully explore their therapeutic potential in managing inflammation and related conditions.

The DPPH assay was utilized to assess the scavenging activity of the tested plant extracts alongside the standard antioxidant, Torolox. This assay determines the concentration of the extract needed to reduce 50% of the initial DPPH radical. Among the plant extracts, J. candicans demonstrated the highest DPPH free radical scavenging activity, with a notably low IC₅₀ value of 71.97 μg/mL, surpassing C. tubulosa, which had an IC₅₀ value of 97.44 μg/mL. In contrast, M. ciliata and T. hirsute exhibited the lowest scavenging activity, with IC₅₀ values of 156 μg/mL and 193 μg/mL, respectively. For comparison, the reference antioxidant Torolox showed an IC₅₀ value of 56.82 μg/mL. These findings highlight that while J. candicans and C. tubulosa display notable antioxidant activity, Torolox remains the most effective agent for scavenging DPPH radicals, outperforming the plant extracts.

Data collected over the study period showed that animals in the control group exhibited significant weight loss and clinical signs of illness, which are indicative of severe septicemia. In contrast, animals receiving treatment with the combination of desert plant extracts and amikacin demonstrated less pronounced weight loss and a notable improvement in clinical signs, suggesting a better overall health status. Specifically, animals treated with the combination showed an average weight loss of approximately 5% compared to 15% in the amikacin-only group, indicating enhanced recovery with the extract. Additionally, the survival rate in the combination treatment group was higher, with 80% of animals surviving until the end of the study, compared to 50% in the amikacin group alone.

Histopathological examination of lung tissue from unchallenged untreated groups ( Figure 3A ) compared to the challenged untreated groups infected with A. baumannii ( Figure 4a ), E. coli ( Figure 4b ), K. pneumoniae ( Figure 4c ), and P. aeruginosa ( Figure 4d ) revealed significant pathological changes. These included collapse of alveolar walls, emphysema, congestion, hemorrhage/edema, as well as pulmonary vascular alterations such as neutrophil infiltration and thrombus formation, with the highest severity scores observed. In the group infected with A. baumannii, amikacin alone, as shown in Figure 5a , led to moderate alveolar wall congestion (++), whereas the combination therapy, as illustrated in Figure 6a , resulted in a mild severity (+), suggesting a potential ameliorative effect of the plant extract. Similarly, interstitial neutrophil infiltration was moderately severe with amikacin treatment (++), but the combination therapy reduced this to a mild response (+). Alveolar edema and hemorrhage followed a comparable pattern, with amikacin showing moderate severity (++) and the combination therapy reducing this to a mild level (+), indicating a reduction in overall lesion severity. Damage to pneumocytes I and endothelial cells was moderately severe with amikacin (++), but notably absent (-) with the combination treatment, suggesting protective benefits of J. candicans extract against cellular damage. Severe thrombosis (+++) within the alveolar vasculature was observed with amikacin alone, but this was significantly reduced to mild (+) with the combined treatment. Furthermore, alveoli filled with granular necrotic debris and minimal cellular reaction were moderately severe (++) with amikacin, but this decreased to a mild severity (+) with the combination treatment, and a similar trend was noted in alveolar collapse, with scores reduced from moderate (++) to mild (+) in the combination group ( Table 4 ).

For mice challenged with E. coli, the lesions were generally more severe, with amikacin treatment alone ( Figure 5B ) resulting in marked alveolar wall congestion (+++), extensive interstitial neutrophil infiltration (+++), and severe pneumocyte and endothelial cell damage (+++). However, the combination of amikacin and J. candicans extract ( Figure 6b ) substantially mitigated these effects, reducing alveolar wall congestion and interstitial neutrophils and pneumocyte damage to mild (+). This reduction in severity extended to alveolar vasculature thrombosis, where the combination treatment decreased lesion scores from severe (+++) to mild (++), and alveolar collapse from (+++) to (+), indicating a significant reduction in overall damage when using the combined treatment ( Table 4 ).

For K. pneumoniae-challenged mice, treatment with amikacin alone ( Figure 5c ) resulted in interstitial neutrophil infiltration with uniformly severe lesions (+++). However, the combination treatment ( Figure 6c ) significantly reduced these scores to mild levels. Pneumocyte damage, which was moderate in the amikacin-treated group, was further reduced to mild levels in the combination therapy group. Despite these improvements, alveolar edema and hemorrhage exhibited severe lesions (+++) with both treatments. Additionally, the presence of alveoli filled with granular necrotic debris, as well as alveolar wall congestion, thrombosis, and collapse, remained consistently moderate (++) for both treatment options, indicating a persistent level of damage ( Table 4 ).

In mice challenged with P. aeruginosa, amikacin treatment alone ( Figure 5d ) led to moderate lesions across various parameters, including alveolar wall congestion, interstitial neutrophil infiltration, alveolar collapse, and pneumocyte damage, all with moderate scores (++). The combination therapy with J. candidus extract ( Figure 6d ) did not markedly alter these outcomes, as most parameters remained at similar moderate levels (++). The severity of alveoli filled with granular necrotic debris was mild (+) in both treatment groups, showing minimal variation. However, a notable improvement was observed in pneumocyte damage with the combination treatment, which was completely resolved, resulting in a healing score (-). Additionally, the combination therapy reduced alveolar thrombosis severity from severe (+++) in the amikacin treated group to moderate (++). Alveolar edema and hemorrhage also improved, from moderate lesions with amikacin treatment to mild lesions with the combination therapy. These findings suggest that while the combination therapy significantly enhanced pneumocyte healing and reduced Alveolar edema, hemorrhage, and thrombosis severity, other aspects such as lung interstitial infiltration, alveolar collapse, pneumocyte damage, and necrotic debris were less responsive to the combined treatment approach ( Table 4 ).

In the histopathological study comparing splenic tissue from unchallenged untreated mice (control negative) shown in Figure 3b with that from untreated challenged mice ( Figure 7 ) infected with A. baumannii ( Figure 7a ), E. coli ( Figure 7b ), K. pneumoniae ( Figure 7c ), and P. aeruginosa ( Figure 7d ), the challenged mice exhibited a high severity score with significant pathological changes. There was a notable increase in apoptosis and necrosis of lymphocytes, indicating substantial cellular damage and death within the spleen. Furthermore, the presence of considerable fibrosis and/or hyaline deposition in the splenic tissue points to chronic inflammation and tissue remodeling. These results highlight the severe impact of the infections on the spleen, showcasing extensive tissue damage and a compromised immune response. The splenic score lesion analysis of mice treated with amikacin alone ( Figure 8 ) or in combination with J. candicans extract ( Figure 9 ) showed notable differences in the severity of apoptosis/necrosis of lymphocytes and fibrosis/hyaline deposition across the different bacterial infections. For mice challenged with A. baumannii, treatment with amikacin ( Figure 8a ) resulted in mild lymphocyte apoptosis/necrosis (+), whereas the combination with J. candicans extract ( Figure 9a ) showed no detectable apoptosis/necrosis (-), suggesting a protective effect of the plant extract. Additionally, there was no fibrosis or hyaline deposition observed in either the amikacin alone or the combination treatment groups (-), indicating minimal splenic tissue remodeling in response to the infection or treatment. In the case of E. coli infection, amikacin treatment ( Figure 8b ) led to moderate lymphocyte apoptosis/necrosis (++), while the combination with J. candicans ( Figure 9b ) significantly reduced this to mild levels (+), further supporting the potential protective role of the plant extract against splenic damage. No fibrosis or hyaline deposition was detected in either treatment group (-), suggesting that the treatments did not induce significant fibrotic changes within the spleen ( Table 4 ).

For K. pneumoniae infections, neither amikacin alone ( Figure 8c ) nor the combination with J. candicans extract ( Figure 9c ) resulted in detectable lymphocyte apoptosis/necrosis or fibrosis/hyaline deposition (-), indicating a lack of observable splenic lesions with these treatments. This suggests that the bacterial strain or the specific conditions did not provoke splenic damage under these treatment regimens. In mice challenged with P. aeruginosa, treatment with amikacin alone ( Figure 8d ) resulted in severe lymphocyte apoptosis/necrosis (+++), highlighting a significant degree of splenic damage. However, the combination with J. candicans extract ( Figure 9d ) markedly reduced this severity to mild levels (+), demonstrating a substantial protective effect. For fibrosis and hyaline deposition, amikacin alone showed mild changes (+), whereas the combination treatment showed no signs of fibrosis or hyaline deposition (-), further indicating that the extract may help mitigate fibrotic processes in the spleen ( Table 4 ).

Overall, the inclusion of J. candicans extract with amikacin demonstrated a general trend of reduced lesion severity across several parameters, particularly in cases involving A. baumannii and E. coli, where notable reductions were observed in alveolar congestion, neutrophil infiltration, and cellular damage. This combination therapy appeared to enhance the therapeutic effects of amikacin, reducing the severity of pulmonary sepsis. However, its efficacy varied depending on the specific pathogen, as seen with K. pneumoniae and P. aeruginosa, where the combination showed limited or no additional benefit in terms of reducing lesion severity. In the spleen, the combination therapy consistently led to reduced severity of lymphocyte apoptosis/necrosis and prevented fibrosis or hyaline deposition in most bacterial challenges, especially with A. baumannii, E. coli, and P. aeruginosa. These findings highlight the potential of J. candicans extract to enhance the protective effects of amikacin, offering additional support in treating bacterial infections associated with significant alveolar, vascular, and splenic damage, although its effectiveness may depend on the specific pathogen involved.

This study presents a novel approach to tackling multidrug-resistant (MDR) bacterial infections by exploring the antimicrobial potential of desert-adapted medicinal plant extracts. The integration of microbiological, biochemical, and histopathological analyses provides a comprehensive evaluation of the therapeutic properties of these extracts. A key strength is the discovery that J. candicans extract enhances the efficacy of amikacin, converting resistant bacterial strains into sensitive ones, which holds significant clinical implications. Additionally, the extract exhibits anti-inflammatory (COX-1 and COX-2 inhibition) and antioxidant properties, making it a promising dual-action therapeutic agent. The study also directly addresses the growing crisis of antibiotic resistance in septic patients, an area of high clinical relevance, and demonstrates improved treatment outcomes in an in vivo model. The findings suggest that plant-based compounds could serve as adjunct therapies, supporting their potential in future drug development.

Despite its strengths, the study does not investigate the precise molecular mechanisms underlying the antimicrobial and anti-inflammatory effects of the extract, which could provide deeper insights into its mode of action. While the in vivo study in mice is a valuable step, larger preclinical and clinical trials are necessary to confirm safety and efficacy in humans. Another limitation is that the long-term impact on bacterial resistance evolution was not assessed, which is critical for determining the sustainability of plant-based adjunct therapies. Lastly, while histopathological analysis indicated reduced tissue damage, detailed toxicity studies are required to ensure the extract’s long-term safety for clinical applications.