1. Introduction The global rise in antimicrobial resistance (AMR) poses a significant public health challenge, accounting for about 9% of annual deaths worldwide and significantly contributing to healthcare-associated morbidity and mortality [ 1 ]. Klebsiella pneumoniae, a gram-negative pathogen, is one of the six major causes of hard-to-treat hospital-acquired infections; it can rapidly develop resistance to multiple classes of antibiotics [ 2 , 3 ].
In response, the World Health Organization have prioritized the development of novel antimicrobial strategies to address this escalating threat [ 4 , 5 ]. However, developing new antibiotics is a long and costly process, often taking 10–15 years, which poses a challenge given the rapid pace of AMR. In this context, antibiotic combination therapy offers a suitable alternative, since the drugs involved are well-known [ 6 ].
When used synergistically, combination treatments can broaden the antibacterial spectrum, increase bacterial killing, decrease the chances of developing resistance, and lower the required doses of individual drugs, thereby minimizing toxicity [ 7 – 9 ]. They can also target multiple bacterial pathways simultaneously, making it more difficult for pathogens to adapt and survive; thus, they are particularly effective against multidrug-resistant (MDR) strains [ 10 ]. Cefotaxime (CTX), a third-generation cephalosporin, is widely used as a broad-spectrum antibiotic [ 11 ].
Although β-lactam resistance mechanisms, especially those driven by plasmid-mediated extended-spectrum β-lactamases (ESBLs), have emerged, CTX remains effective against many bacterial strains in clinical practice [ 12 , 13 ]. To address these challenges and improve treatment outcomes, combination strategies involving CTX and β-lactamase inhibitors (e.g., sulbactam) [ 14 ], or other synergistic antimicrobials (e.g., rifampicin, ciprofloxacin, delafloxacin) are being explored to enhance efficacy against resistant strains [ 15 , 16 ]. A synergistic interaction between CTX and polymyxin B (PMB) was observed in a recent study against a multidrug-resistant, polymyxin-carbapenem resistant (PC-R) K. pneumoniae strain [ 17 ].
While the toxicity profiles of CTX and PMB mono-therapies are available, their combined use, denoted here as CTX/PMB, has not been thoroughly investigated. Preclinical studies are needed to ensure the safety and efficacy of new treatment strategies, serving as an important step toward real-world application [ 18 ]. Therefore, this study evaluated CTX/PMB in a mouse model by assessing both potential toxicity indices by performing an extended single-dose assay and determining antimicrobial activity against PC-R K. pneumoniae in a mammalian system.
2. Material and methods 2.1 Chemicals CTX sodium salt (Lot #0000334303) and PMB solution at a concentration of 1 mg/mL (Lot #BCCG2613) were purchased from Sigma-Aldrich. CTX was prepared according to the manufacturer’s instructions, being diluted in saline solution for use in animal experiments.
2.2 Animals BALB/c mice (8–12 weeks old, 24–28 g) were obtained from the Central Animal Facility of the Federal University of Mato Grosso do Sul. Prior to experimentation, the animals underwent a one-week acclimatization period. The mice were housed in polypropylene cages under controlled conditions: temperature (22 ± 3 °C), humidity (40–60%), and a 12 h/12 h light-dark cycle.
Standard commercial feed and water were provided ad libitum . All procedures were performed between September and November 2024 in accordance with guidelines established by the National Council for the Control of Animal Experimentation (CONCEA). Further, the study protocol was reviewed and approved by the Animal Ethics Committees of the Federal University of Grande Dourados (protocol no.
23018) and the University Center of Grande Dourados – UNIGRAN (protocol no. 080/18). Manipulations were performed by personnel who had undergone prior training in rodent handling, under the guidance of senior staff with extensive experience in mouse experimentation.
Humane endpoints were predefined in accordance with institutional animal care standards, including criteria such as >20% body weight loss, persistent immobility, inability to access food or water, or signs of severe pain or distress. Throughout the study, no animals met these humane endpoint criteria. 2.3 Extended single-dose toxicity assay An extended single-dose toxicity assay was conducted to evaluate the toxicity of CTX/PMB [ 19 ].
Briefly, the animals were randomly divided into four groups, each consisting of six mice. The mice were administered a single intraperitoneal (i.p.) injection of the following treatments: (1) Control: vehicle only (saline solution); (2) Low-dose: CTX/PMB with CTX at 2 g; (3) Medium-dose: CTX/PMB with CTX at 4 g; (4) High-dose: CTX/PMB with CTX at 8 g ( Table 1 ). The low CTX dose was based on standard human posology (assuming a 70 kg adult) for uncomplicated infections [ 20 ] and adapted for mice using body weight scaling.
The medium and high doses corresponded to two times and four times the standard dose, respectively. The PMB concentration was fixed at 2 mg/kg across all combination groups, corresponding to its standard posology [ 21 ]. 10.1371/journal.pone.0339990.t001 Table 1 Treatment groups and survival rates in the extended single-dose toxicity assay for CTX/PMB.
Group Treatment dosage Survival (%) Cefotaxime g/day* Polymyxin B mg/kg/day Control 0 0 100 Low-dose 2 2 100 Mid-dose 4 2 100 High-dose 8 2 100 *Doses were derived from human therapeutic regimens (70 kg reference weight) and translated to mice using body weight-based scaling. 2.3.1 Animal monitoring. After treatment, the mice were monitored daily for 14 days to assess their general condition, including behavior, appearance, physiological function, presence of convulsions, pruritus, ataxia, and body weight [ 22 ].
Food consumption was evaluated by measuring the daily reduction from a standard 0.2 kg portion. Water consumption was evaluated using the same approach, with an initial volume of 0.6 L of water provided per cage. Body weight, food intake, and water consumption over time were analyzed by calculating the area under the curve (AUC) using the trapezoidal rule [ 23 ].
2.3.2 Blood collection and organ weight measurement. After monitoring for 14 days, the mice were anesthetized via i.p. injection of a combination of xylazine (10 mg/kg) and ketamine (60 mg/kg). Euthanasia was performed by exsanguination under deep anesthesia.
Blood samples collected during this procedure were analyzed for biochemical parameters. Then, the mice were dissected following standard pathological procedures. The liver, spleen, and both kidneys were excised and weighed [ 24 ].
2.3.3 Biochemical analyses. For biochemical analyses, plasma was obtained by centrifuging the blood sample collected in tubes, without anticoagulant, at the exsanguination procedure [ 25 ]. An automated clinical chemistry analyzer, Mindray BS-120 (Shenzhen Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China), was used to perform biochemical serum analyses.
The parameters evaluated included creatinine, urea, alanine aminotransferase, alkaline phosphatase, aspartate aminotransferase, total bilirubin, direct bilirubin, and indirect bilirubin. All analyses were conducted under standardized laboratory conditions, and the results were interpreted based on reference values. 2.4 Infection model 2.4.1 Bacterial strain and culture conditions.
To evaluate the antibacterial activity of CTX/PMB, an infection model was established using the PC-R K. pneumoniae K18 strain, which was previously described exhibiting resistance to carbapenems, β-lactams, and polymyxins [ 26 ]. For the experiment, the isolate was initially cultured in brain heart infusion (BHI) broth at 37 °C for 24 h. Following this incubation, the culture was streaked onto BHI agar and incubated under the same conditions at 37 °C for 24 h.
This protocol ensured the optimal growth and viability of PC-R K. pneumoniae K18 for the subsequent assay. 2.4.2 In vivo antibacterial activity evaluation. Briefly, dexamethasone (20 mg/kg) was administered via the i.p. route to the mice 24 h before the experiment to induce immunosuppression and increase model sensitivity.
All neutropenic mice, except those in the naïve group, were administered an i.p. injection of 0.2 mL of bacterial suspension containing 3 × 10 8 CFU/mL, prepared in 0.9% saline solution [ 27 ]. The mice were randomly assigned to five groups as follows: (1) naïve: uninfected control (n = 4); (2) mock-treated: infected but untreated control (n = 5); (3) CTX: received CTX 2 g/kg/day (dose adjusted to mouse body weight) i.p. (n = 4); (4) PMB: received PMB 1 mg/kg/day i.p.
(n = 4); and (5) CTX/PMB: received both CTX (2 g/day) and PMB (1 mg/kg/day) i.p (n = 4). The CTX dose was selected based on the results of the toxicity assay, whereas the PMB dose was determined according to its pharmacokinetic profile; specifically, the plasma concentration required over time to reach its effective in vitro dose in an MDR K. pneumoniae infection model [ 28 ]. Animals were monitored every 6 h for well-being and mortality.
After 24 h, the mice were anesthetized via i.p. injection of xylazine (10 mg/kg) and ketamine (60 mg/kg). Euthanasia was performed by exsanguination under deep anesthesia. Blood samples collected at this stage were plated on BHI agar containing meropenem to selectively recover resistant bacteria.
After 24 h of incubation at 37 °C, bacterial colonies were counted and expressed as log 10 CFU/mL, with a lower detection limit set at 1.69 log 10 CFU/mL. 2.5 Statistical analyses Statistical analyses were performed using One-way analysis of variance (ANOVA) followed by Tukey’s test for group comparisons. Survival of infected mice was evaluated using Kaplan–Meier survival curves.
Differences were considered statistically significant at p < 0.05. All analyses were conducted using GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA). 3.
Results 3.1 Extended single-dose toxicity assay 3.1.1 Animals general condition. Although CTX and PMB have known safety profiles, the effects of their combination are not known. In the groups treated with CTX/PMB, which comprised three different doses of CTX along with a fixed dose of PMB, no deaths or visible abnormalities were recorded, as was the control group (vehicle only).
All treated mice showed 100% survival ( Table 1 ). Throughout 14 days of monitoring, the animals remained socially active, with no apparent changes in behavior or general condition. 3.1.2 Animals body weight and food consumption.
Throughout 14 days of monitoring, the medium-dose and high-dose groups maintained stable body weights, showing no significant increasing trends, similar to those of the control group. However, the low-dose group presented a lower average body weight from the start of the experiment, with only a slight reduction of 0.5 g by the final day ( Fig 1A.1 ). This difference was statistically significant compared to the control group, as determined by the analysis of the weight/time AUC ( Fig 1A.2 ).
10.1371/journal.pone.0339990.g001 Fig 1 Mice body weight (A.1) was monitored over 14 days following a single dose of CTX/PMB at three different CTX concentrations (low, mid, and high) or vehicle (control). The area under the weight curves was calculated for each group (A.2) . Food consumption was recorded daily for all groups (B.1) , and corresponding areas under the food consumption curves were obtained (B.2) .
Water consumption was monitored over the same period (C.1) , and the area under the curves was calculated for each group (C.2) . Area under the curve (AUC) values were analyzed for statistical significance using one-way ANOVA, with p < 0.05 considered significant. Food consumption was assessed by measuring the reduction in a standard portion over time; thus, the area under the plotted curves was inversely proportional to actual intake.
All groups consumed similar amounts of food throughout the experiment ( Fig 1B.1 ). Water consumption was evaluated using the same method, with comparable results recorded across groups ( Fig 1C.1 ). The low-dose group presented slightly lower food and water intake (as shown in AUC analyses), which reflected the lower median body weight of the mice in this group ( Fig 1B.2 and 1C .2).
3.1.3 Mice organ weights. Organ weights were measured to assess potential effects of single-dose CTX/PMB administration ( Table 2 ). The low-dose group presented the lowest values across all organs assessed, which was probably attributable to their overall lower body weight.
Medium-dose group did not have any significant difference in organs weight. However, compared to the control (vehicle only), the high-dose group presented statistically significant reductions (p < 0.05) in the right kidney and liver. 10.1371/journal.pone.0339990.t002 Table 2 Mice organ weights 14 days after receiving a single dose of CTX/PMB or vehicle.
Organ Control Cefotaxime/Polymyxin B Low-dose Mid-dose High-dose Spleen 0.126 ± 0.007 0.113 ± 0.013 0.119 ± 0.012 0.118 ± 0.017 Right Kidney 0.189 ± 0.016 0.165 ± 0.011 0.172 ± 0.012 0.165 ± 0.013 Left Kidney 0.187 ± 0.017 0.157 ± 0.012 0.170 ± 0.019 0.167 ± 0.017 Liver 1.321 ± 0.152 1.075 ± 0.126 1.207 ± 0.096 1.100 ± 0.081 All values are expressed as mean ± standard error and reported in g. 3.1.4 Biochemical results. Biochemical analysis revealed no statistically significant differences among the groups for any of the evaluated parameters.
However, mice treated with CTX/PMB at low and medium doses presented high mean levels of creatinine, which exceeded the upper reference limit (indicated by the dashed line). A slight increase in creatinine was also observed for one animal in the control group ( Fig 2A ). Additionally, the high-dose group presented high urea levels ( Fig 2B ).
These variations suggested a response to PMB. All other biochemical parameters were within normal reference ranges ( Fig 2C – 2H ). 10.1371/journal.pone.0339990.g002 Fig 2 The biochemical markers in mice treated with CTX/PMB at three different CTX doses (low, medium, and high) and the control group (vehicle only) were quantified.
The parameters included A) creatinine, B) urea, C) alanine aminotransferase (ALT), D) alkaline phosphatase, E) aspartate aminotransferase (AST), F) total bilirubin, G) direct bilirubin, and H) indirect bilirubin. The dashed lines indicate the upper reference limits for each parameter. Statistical analysis was performed via one-way ANOVA; no significant differences were found among the groups.
3.2 Infection model Based on previous evidence of the synergistic activity of CTX/PMB against PC-R K. pneumoniae and the results of the extended single-dose toxicity assay, a murine infection model was constructed to evaluate whether this effect is maintained in a complex mammalian in vivo system ( Fig 3A ). After 24 h, the CTX-treated mice exhibited a survival rate of only 25%, which was lower than that of the mock-treated group (survival rate of 40%). In contrast, all animals in the CTX/PMB-treated, PMB-treated, and naïve groups survived during this period ( Fig 3B ).
The CTX/PMB-treated mice were more active throughout the experiment, suggesting higher treatment efficacy. 10.1371/journal.pone.0339990.g003 Fig 3 A) Schematic representation of the PC-R K. pneumoniae K18 strain infection model in mice used to evaluate the antimicrobial efficacy of the CTX/PMB combination against a resistant strain. B) Survival curves of mice treated with PMB, CTX, or the CTX/PMB combination following PC-R K. pneumoniae K18 infection.
C) Bacterial load (log 10 CFU/mL) across treatment groups. A lower detection limit of 1.69 log 10 CFU/mL was established. Naïve (uninfected) and mock-treated (infected but untreated) control groups were included.
Statistical analysis was performed via one-way ANOVA; the differences among groups were considered to be statistically significant at p < 0.05. To determine the effectiveness of the treatments against PC-R K. pneumoniae K18 strain infection, the bacterial load was measured in mouse blood samples. The CTX/PMB regimen significantly decreased bacterial levels, reaching values comparable to those of the naïve (uninfected) group.
However, mice treated with either antibiotic alone did not show the same improvement, with bacterial loads remaining similar to those in the mock-treated (infected and untreated) group ( Fig 3C ). 4. Discussion The AMR has become a growing threat to global public health.
To tackle this challenge, the continued development of effective therapies is crucial, with combination treatments emerging as cost-effective and timely solutions 4 . CTX/PMB has shown strong synergistic activity against PC-R K. pneumoniae , demonstrated by a low fractional inhibitory concentration index (FICI < 0.5) and a high zero interaction potency (ZIP) synergy score of 37.484. The combination also showed no hemolytic activity (hemolysis rate < 5%) and no detectable toxicity in the alternative nematode model Caenorhabditis elegans [ 17 ].
These promising results encouraged this study to advance to preclinical testing in mice, as no such study has been reported to date, marking a vital step toward translating laboratory findings into clinically applicable treatment strategies. The cyclic cationic polypeptide antibiotic, PMB, is considered an important last-resort treatment for MDR Gram-negative infections [ 26 ]. However, its clinical utility is limited by significant adverse effects, particularly neurotoxicity and nephrotoxicity, the latter often resulting in acute kidney injury characterized by reduced glomerular filtration rate and elevated serum levels of metabolic waste products such as creatinine and urea [ 29 ].
Interestingly, less toxic PMB derivatives are currently being study and might represent a promising solution for this problem [ 30 , 31 ]. The present study evaluated whether CTX, administered at three different doses in combination with a fixed dose of PMB, would influence or worsen the toxicity profile of PMB to develop safe CTX/PMB regimens. Single-dose administration of CTX/PMB did not result in mortality or observable adverse effects in mice.
Throughout 14 days of monitoring, all animals remained active and showed no signs of distress. Additionally, the combination treatment did not significantly affect body weight, food intake, or water consumption in the medium-dose or high-dose CTX groups. The maintenance of stable body weight in toxicity studies is considered to be an early indicator of general well-being, systemic safety, and preservation of organ integrity in preclinical models [ 32 ].
Therefore, the absence of weight loss in treated groups supports a preliminary safety profile of CTX/PMB. After 14 days of monitoring, the mice were euthanized for organ weight assessment and biochemical analysis. The group treated with the higher dose of CTX in combination with PMB showed a significant reduction in the weight of the right kidney and liver, accompanied by an increase in plasma urea levels.
The combination-treated groups also presented creatinine levels at the upper limit; however, a similar increase was noted in the control group, indicating that this effect may not be solely related to treatment. These results suggested a dose-dependent toxicity pattern associated with CTX, which may be amplified by the presence of PMB. Considering the well-known nephrotoxic profile of PMB, typically marked by increased serum levels of creatinine and urea [ 33 ], the findings of the present study highlight a possible pharmacotoxicological interaction of the combination on renal function at higher CTX doses, which should be carefully considered while developing clinical dosing regimens.
Polymyxins (PMB and colistin) exhibit neurotoxic effects through inhibition of acetylcholine release and mitochondrial toxicity. The neurotoxic effects of PMB are characterized by signs such as ataxia, convulsions, and pruritus [ 28 , 34 ]. The absence of these signs in the treated animals suggests that neurotoxic effects were not manifested at the administered dose of 2 mg/kg.
Based on the results and the indication of a dose-dependent toxic effect associated with CTX, the medium dose of CTX was selected for subsequent infection in an in vivo model. This was combined with PMB at 1 mg/kg/day, a dose selected according to its pharmacokinetic profile in the treatment of PC-R K. pneumoniae infections [ 28 ]. Compared to mono-therapies, CTX/PMB demonstrated enhanced antibacterial activity, significantly reducing the blood bacterial load of the PC-R K. pneumoniae in treated neutropenic mice while also prolonging overall survival and improving clinical conditions.
Similar to the findings of the present study, other studies have shown that PMB exhibits synergistic effects with rifampicin against Gram-negative pathogens in neutropenic mouse models, reinforcing that PMB-based combination strategies may be effective in treating MDR infections [ 35 , 36 ]. From a clinical perspective, the translation of CTX/PMB therapy requires careful evaluation of dosing strategies to ensure optimal therapeutic outcomes. Given the pharmacokinetic properties of both agents, dose adjustment and therapeutic monitoring may help maintain efficacy while minimizing adverse effects (such as nephrotoxicity) during prolonged treatment, especially in critically ill patients [ 37 , 38 ].
Importantly, CTX/PMB represents a promising combination for clinical application, as both agents are already established as effective mono-therapies and are widely available for clinical use. This study has some limitations. The toxicity assessment was based on a single-dose model, which may not reflect potential long-term effects, such as cumulative renal and neurological injury caused by antibiotics exposure.
Further, the infection model used only one strain of PC-R Klebsiella pneumoniae , limiting the generalization of the findings to other strains or resistance profiles. Plus, the group sample size was small, which may have resulted in less robust statistical outcomes. The findings were based on a murine model, which may not fully reflect human responses.
Finally, the observed increase in creatinine and urea suggests potential nephrotoxicity that requires a deeper investigation. Future studies should explore repeated-dose models, diverse bacterial strains, and advanced renal and neurotoxic assessments. 5.
Conclusion In conclusion, the CTX/PMB regimen demonstrated promising antibacterial efficacy against PC-R K. pneumoniae in a murine model, significantly reducing bacterial load and improving survival. These findings suggest that while CTX/PMB has therapeutic potential, careful optimization of dosing is essential to balance efficacy and safety, making further studies essential to establish the optimal therapeutic regimen and fully characterize the safety profile of this combination.