Introduction

Urinary tract infection (UTI) is a highly prevalent infectious disease in the global population, impacting an estimated 150 million individuals annually. This condition poses substantial economic and public health challenges [1, 2]. Among the various causative agents, Escherichia coli, Pseudomonas aeruginosa, Staphylococcus aureus, Klebsiella pneumoniae, and Candida species are the most frequently encountered pathogenic bacteria associated with UTIs. These bacteria possess the capability to persist within urothelial cells, and inadequate treatment can result in recurrent UTI episodes, thereby elevating the risk of secondary bloodstream infections [3].

The conventional treatment approach for urinary tract infections involves selecting beta-lactam agents, fluoroquinolones, trimethoprim-sulfamethoxazole, amoxicillin-clavulanic acid, or third-generation cephalosporins based on bacterial culture sensitivity results [4]. However, the treatment of urinary tract infections is faced with challenges due to the emergence of drug-resistant bacteria and the formation of intracellular bacterial reservoirs [5]. Additionally, the utilization of antimicrobial agents in this context is also limited by adverse drug reactions and drug interactions [6]. Moreover, a growing body of research indicates that antimicrobial drugs have the potential to disturb the ecological equilibrium of the gut microbiota and potentially enhance bacterial infections through modulation of the immune response [3, 52], and more, in material synthesis. The combined application of these compounds undoubtedly enhances the functionality of materials and expands their potential applications. To further investigate the impact of Lev@PADM’s long-lasting antimicrobial effect on the innate immune response, we first measured the body weight and organ weight of mice in each group before and after infection. Results showed a significant decrease in body weight and a notable augmentation in bladder, spleen, and kidney weight. Solely the mice treated with Lev@PADM demonstrated a substantial decrease in organ weight, subsequently restoring them to their baseline levels (Fig. 7A, B). This observation suggests that the administration of Lev@PADM effectively mitigated the enlargement of the bladder, kidney, and spleen induced by bacterial infection. Immunohistochemistry (IHC) staining was performed for neutrophils, M1 macrophages, M2 macrophages, and mast cells, which are prominent innate immune cells in the tissue (Fig. 7C, D). The findings revealed a significant increase in the quantities of neutrophils, M1 macrophages, and M2 macrophages in the bladder tissue of infected mice after a duration of seven days, potentially linked to the persistent presence of E.coli in the bladder (Fig. 6B). The bladder treated with Lev@PADM exhibited a significant decrease in the quantities of neutrophils, M1 macrophages, M2 macrophages, and mast cells compared to the levofloxacin-treated group. This suggests that the Lev@PADM hydrogel not only extends the antimicrobial efficacy of levofloxacin and diminishes bacterial retention in the bladder, but also suppresses excessive activation of the innate immune response.

Fig. 7
figure 7

The impact of Lev@PADM on the innate immunity in cases of bacterial cystitis. A Variations in body weight prior to and following the administration of saline, Levofloxacin, PADM, and Lev@PADM in a bacterial cystitis model. B Weight of kidneys, bladder, and spleen in mice belonging to different treatment groups after 7 days of infection. C Expression of iNOS, CD206, MPO, and Chymase in bladder tissue on the 7th day post- treatment. D Optical density analysis of positive cells in C. (*p < 0.05; **p < 0.01; ***p < 0.001; ns: no significant. Scale bar: 200 μm. Black boxes indicate representative areas of positive protein expression)

Full size image

In order to further investigate the mechanism by which Lev@PADM inhibits innate immunity, we performed Western blot (WB) analysis on treated bladder tissue to examine the levels of TLR4, p-P65, p-IκB, TNFα, and IL-1β (Additional file 4: Figure S4A, B). Extensive research has shown that E. coli can activate the NF-κB signaling pathway by binding to TLR4, promoting the phosphorylation and nuclear translocation of P65 (Additional file 5: Figure S5A, B), and subsequently transcribing TNFα and IL-1β to promote immune cell infiltration and activation. However, excessive immune response can also lead to barrier disruption, thereby promoting E. coli invasion of the urinary tract. Therefore, mitigating NF-κB signaling activation, while controlling bacterial infection, is beneficial for the restoration of the bladder epithelial barrier. Our results revealed a significant upregulation of the NF-κB signaling pathway in bladder tissues after infection. Treatment with levofloxacin inhibited NF-κB activation, and compared to levofloxacin, Lev@PADM further suppressed NF-κB activation. This suggests that Lev@PADM treatment may potentially reduce NF-κB activation, decrease the expression of inflammatory factors TNF-α and IL-1β, and subsequently diminish immune cell infiltration, which may be the main reason for Lev@PADM’s reduction in immune cell infiltration. Consequently, this advantageous outcome contributes to the preservation of the bladder epithelial barrier’s integrity and facilitates the recuperation from urinary tract infections.

Maintenance of gut microbiota structure

The perturbation of the gut microbiota resulting from the administration of antimicrobial agents is linked to dysregulation of the host immune system, thereby heightening vulnerability to diverse bacterial infections [53, 54]. Prior investigations have unequivocally established that the modification of the gut microbiota composition brought about by levofloxacin intensifies the inflammatory reactions of the host [55]. This implies that mitigating the influence of levofloxacin on the gut microbiota holds considerable significance in the management of urinary tract infections.

In this study, we conducted 16S rRNA sequencing on fecal samples obtained from three groups of mice: the control group, the levofloxacin oral gavage group, and the Lev@PADM bladder instillation group. To identify species and genes, Operational Taxonomic Units (OTUs) were clustered at a 97% similarity level. The OTU species accumulation curve reached a plateau as the sample size increased, suggesting that the sample size was sufficient for the experimental requirements (see Fig. 1A). Additionally, the rank-abundance curve indicated a wider range of species abundance in the Lev@PADM group compared to the levofloxacin group, suggesting a higher species abundance in the Lev@PADM group. Furthermore, the α-diversity analysis of the gut microbiota, specifically Chao1 and Observed-species metrics, revealed no statistically significant disparities in species abundance between the Lev@PADM group and the control group. Conversely, the levofloxacin oral gavage group displayed noteworthy alterations in species abundance when compared to the control group, as depicted in Fig. 7C. The proximity of scatter points in the PCA plot indicates minimal variations within the intra-group samples, thereby implying the experiment’s satisfactory repeatability and reliability. The samples obtained from the group administered with levofloxacin orally and the control group were observed to be distinct from each other, implying that oral gavage of levofloxacin resulted in alterations in the composition of gut microbiota. Notably, the samples acquired from the Lev@PADM bladder instillation group exhibited a closer resemblance to the control group. This observation indicates that bladder instillation of Lev@PADM offers superior preservation of the structure of gut microbiota in comparison to oral administration of levofloxacin, thereby emphasizing the potential of bladder instillation with Lev@PADM as a therapeutic approach for urinary system infections (Fig. 8).

Fig. 8
figure 8

Intravesical infusion of Lev@PADM effectively prevents disruption of the gut microbiota. A Accumulation curve of OTU species. B Rank-abundance curve. C Alpha diversity analysis of gut microbiota. D PCA analysis of gut microbiota composition in different groups. (*p < 0.05.)

Full size image

Discussion

We conducted a comprehensive investigation into the physical properties and biological activity of Lev@PADM hydrogel, which was introduced for the first time in the treatment of bacterial cystitis. Our findings highlight three advantages of utilizing Lev@PADM hydrogel for the treatment of bacterial cystitis. Firstly, the hydrogel exhibits sustained release of levofloxacin and remarkable resistance to deformation. This feature allows for targeted elevation of local levofloxacin concentration at the site of infection through bladder injection, thereby avoiding the hepatotoxic, nephrotoxic, and neurotoxic side effects that may arise from high systemic drug concentrations. Furthermore, the Lev@PADM hydrogel demonstrates favorable biocompatibility and anti-inflammatory characteristics. Consequently, in conjunction with its antimicrobial properties, it effectively sustains the integrity of the bladder epithelium, thereby positively impacting the prognosis of bladder infections. Lastly, when compared to the traditional oral administration of levofloxacin, the intravesical administration of Lev@PADM hydrogel preserves the composition of the indigenous gut microbiota. This aspect holds considerable significance in terms of maintaining the host’s immune function and preventing the recurrence of urinary tract infections.

Hydrogel materials, through the encapsulation of antibiotics, metal nanoparticles, and other antimicrobial agents, have been widely utilized in the treatment of bacterial infections [56]. According to reports, HA-PAR hydrogel has demonstrated persistent antibacterial activity in both in vitro and in vivo experiments by releasing polyarginine over an extended period [19]. Similarly, P(AA10-co-HEMA10)Gen hydrogel exhibits a pH-dependent sustained release of gentamicin for the treatment of skin infections [57]. These studies suggest that the use of antibacterial drug-loaded hydrogels may offer a time advantage for the treatment of bladder infections. Despite the existence of extensive research studies on the utilization of hydrogel systems for intravesical drug delivery [58, 59], the implementation of Lev@PADM in the management of bacterial cystitis presents an innovative intervention. Hydrogels possess the capacity to mitigate drug dilution and confer adhesive or floating characteristics to loaded substances, thereby prolonging the drug’s efficacy within the bladder [60, 61]. A study documented the development of a mucoadhesive hydrogel composed of high molecular weight chitosan and β-glycerophosphate, which exhibited enhanced resistance to urine washout when evaluated ex vivo using porcine urinary bladder, thus presenting a potential treatment option for bladder cancer [62]. In a separate study, floating hydrogels comprised of Poloxamer 407 (P407) were investigated for their efficacy in managing bladder-related diseases [63]. The hydrogel demonstrated prolonged residence time and continuous drug release for up to 10 h when tested in vivo using a rabbit acute bladder injury model. In our investigation, we employed decellularized porcine dermis hydrogel as the carrier for Levofloxacin. This bio-matrix hydrogel endowed Lev@PADM with adhesive properties and gradual release attributes, enabling it to adhere to bladder epithelial cells and furnish a biologically benign E.coli protective barrier that emulates the inherent mucosal surface. Consequently, PADM can collaborate with therapeutic drugs, particularly in instances where the mucosal barrier is compromised.

E.coli is the primary causative bacterium responsible for bacterial cystitis. The lipopolysaccharides (LPS) present on its surface bind to the TLR4 receptors located on the epithelial cells of the bladder, thereby initiating the activation of IKK and facilitating the translocation of NF-κB into the nucleus. This sequential process ultimately leads to the regulation of various pro-inflammatory cytokines, including IL-1β, TNF-α, and IL-6 [9, 64,65,66,67]. Consequently, the activation of Ly6C − macrophages and mast cells occurs, while neutrophils and Ly6C + macrophages are recruited to eliminate uropathogenic E.coli (UPEC) that has invaded the bladder epithelial cells [68]. However, an overabundance of inflammation can also lead to compromised tight junctions, heightened barrier permeability, and deficiencies in umbrella cells. The absence of fully developed epithelial cell coverage renders the basal epithelial cells more vulnerable to UPEC adhesion, biofilm formation, and even intracellular bacterial colonization, resulting in the persistence of bacterial cystitis [69]. Our research findings have shown that, in comparison to levofloxacin, Lev@PADM consistently inhibits the proliferation of E.coli within the bladder and significantly diminishes intracellular bacterial production in bladder tissue. The results of our study indicate that Lev@PADM exhibits prolonged inhibition of E.coli proliferation in the bladder and effectively reduces intracellular bacterial production in bladder tissue, when compared to levofloxacin. These effects can potentially be attributed to the adhesive characteristics of Lev@PADM to the bladder epithelial barrier, its sustained release of levofloxacin, and its ability to suppress excessive immune responses within the bladder.

The intestinal microbiota is widely recognized as the second genome of the human body [70], exerting a pivotal influence on the regulation of host immunity and gastrointestinal function [71]. Previous studies have consistently indicated that the administration of levofloxacin can disturb the equilibrium of the intestinal microbiota, thereby inducing antibiotic-associated inflammation within the gastrointestinal tract [55]. Moreover, previous research conducted by our team and other scholars has provided evidence that the depletion of the intestinal microbiota resulting from the administration of antimicrobial drugs substantially elevates the likelihood of bacterial infectious diseases [53, 54, 72,73,74]. This correlation is closely linked to the reduction of immunoregulatory metabolites derived from the intestinal microbiota. Consequently, it is imperative to closely monitor the stable composition of the intestinal microbiota when employing levofloxacin for the management of urinary tract infections, as it exerts a substantial influence on the prognosis of the infection. In this study, we compared the effects of oral levofloxacin and bladder instillation of Lev@PADM on the intestinal microbiota structure in mice. We found that bladder instillation of Lev@PADM avoided the disruption of the intestinal microbiota caused by levofloxacin. This finding suggests that Lev@PADM, as a biomaterial, holds great potential for the treatment of urinary tract infections.

Conclusion

The immobilization of levofloxacin onto PADM hydrogel through physical adsorption has led to the development of Lev@PADM hydrogel composite, which exhibits controlled drug release properties. This innovative approach offers several advantages for the administration of levofloxacin, as the hydrogel composite possesses excellent resistance to deformation, allowing for its injection into the bladder. Importantly, the synthesis process employed in creating Lev@PADM does not compromise the inherent antibacterial activity of levofloxacin. In fact, our study demonstrates that Lev@PADM exhibits extended antibacterial efficacy compared to levofloxacin alone. This enhanced effectiveness is complemented by its ability to modulate the immune response, thereby aiding in the restoration of the bladder epithelial barrier and preventing bacterial reattachment in urine following infection. Moreover, we have observed that Lev@PADM exerts a reduced impact on the intestinal microbiota compared to levofloxacin alone. This is a significant finding, as it suggests that Lev@PADM could potentially minimize the disruption of the gut microbiome, which is crucial for maintaining overall health. Based on these compelling attributes, Lev@PADM has emerged as a highly promising biomaterial for the management of urinary tract infections. Its slow drug release characteristics, extended antibacterial efficacy, immune regulatory capabilities, and reduced impact on the intestinal microbiota make it a valuable candidate for further exploration and development in the field of urinary tract infection treatment. However, it is important to acknowledge the limitations of our study. Although our experiments demonstrate the potential of Lev@PADM, further research is needed to optimize its formulation, dosage, and long-term safety profile. Additionally, the complex dispersion characteristics of Lev@PADM, compared to levofloxacin alone, warrant careful consideration in future studies.

In conclusion, the immobilization of levofloxacin onto PADM hydrogel to create Lev@PADM represents an innovative approach with significant potential for the management of urinary tract infections. The extended antibacterial efficacy, immune regulatory capabilities, and reduced impact on the intestinal microbiota make Lev@PADM a promising biomaterial for further investigation and development.

Materials and methods

Preparation of Lev@PADM hydrogel

Fresh pig skins were procured from slaughterhouses and subsequently processed by removing soft tissues such as muscle and fascia. These tissues were then sliced into thin sections measuring 2–3 mm in thickness. The slices were placed in beakers and subjected to a three-hour rinsing period at room temperature (25 °C) using sterile water with continuous agitation. Following this, the slices were frozen and thawed for three cycles, with temperatures ranging from − 80 °C to 37 °C. The resulting thin slices of pork skin were immersed in a 1% Triton X-100 (v/v) solution, which was refreshed every 24 h. After a 24-h period, the slices were transferred into a 0.1% Ethylene diamine tetraacetic acid (EDTA) solution and stirred for an additional 24 h. The slices underwent an overnight rinsing process with a substantial quantity of phosphate buffered saline (PBS) to eliminate any remaining chemicals. Subsequently, the treated flakes were subjected to lyophilization using a freeze dryer, and upon obtaining the lyophilized flakes, they were further pulverized into a powder using a grinder. A concentration of 12 mg/ml was introduced into a solution of HCl with a pH range of 2 to 3, which also contained pepsin at a mass ratio of 1:2 with PADM.

powder. The mixed suspension was subjected to continuous stirring for a duration of 24 h at ambient temperature until it achieved homogeneity, resulting in the formation of a flowable pre-gel solution. Subsequently, levofloxacin was incorporated into the pre-gel solution in accordance with the pre-gel volume, thereby obtaining a hydrogel loaded with the levofloxacin. The pre-gel solution was neutralized using a pre-cooled 0.1 M NaOH solution, and the osmotic pressure of the neutralized pre-gel solution was adjusted utilizing PBS. Following the aforementioned procedures, the pre-gel solution was loaded into a syringe, injected into the mold, and left undisturbed at ambient temperature (25 °C) for a duration of 5 min.

Characterisation of hydrogels

The effect of decellularisation was evaluated through the utilization of HE and DAPI staining techniques, which allowed for the observation of decellularisation of the porcine epidermal matrix under a microscope. The hydrogels were subjected to fixation in a 2.5% glutaraldehyde solution at room temperature for a duration exceeding one hour. Subsequently, the samples underwent a series of washing steps, involving three repetitions of a 10-min wash with deionized water. Following this, the samples were subjected to dehydration using ethanol solutions with varying concentrations, namely 30%, 50%, 70%, 80%, 90%, and 95%, with each concentration being applied for a duration of 15 min. Finally, the samples were treated with a 100% ethanol solution and soaked. The specimens were subjected to dehydration using liquid carbon dioxide in a critical point desiccator. Subsequently, the scanning electron microscope (JSM-6360LV, Japan) was employed to examine the structural and surface characteristics of the PADM hydrogel and Lev-PADM hydrogel after sufficient dehydration treatment. The porosity of the cross-sections of both hydrogels was determined by capturing SEM images at a magnification of 10000X, which were then analyzed using Image J software (National Institutes of Health, USA).

Determination of DNA content and collagen content

Small slices of pork skin were cut and a sample weighing 5–25 mg was added to a test tube. After tissue lysis and washing, a DNA extraction kit (TaKaRa, Japan) was used to obtain a purified DNA solution. Subsequently, 100 μl of the solution was added to a 96-well plate, and the OD values were measured at 260 nm using a spectrophotometer to calculate the DNA content. The determination of collagen involved measuring the hydroxyproline content using the chloramine assay, with a hydroxyproline (HYP) to collagen ratio of 1:7.6954. The hydroxyproline concentration standard curve was initially obtained through spectrophotometry using a hydroxyproline ELISA kit. The methodology involved a concise description of the sample’s hydrolysis in 38% concentrated hydrochloric acid at 110 °C for a duration of 18 h, followed by pH adjustment to a range of 6–8. Subsequently, 100 μl of the resulting solution was added to a 96-well plate, and the collagen content was determined by calculating the absorbance at a wavelength of 560 nm using a spectrophotometer.

Fourier-infrared spectroscopy

The infrared spectra of PADM hydrogel, and Lev@PADM hydrogel were analyzed using Fourier-infrared spectroscopy. The examination was conducted through the potassium bromide method, utilizing a Fourier infrared spectrometer (Nicolet Nexus, Thermo, USA) that operates within the 400–4000 cm−1 range.

Rheological analysis

Using a rotational rheometer (MCR302, Anton Paar, Canada), the PADM hydrogel was placed on a parallel plate at 37 °C. The strain was set from 0.1% to 100%, and the frequency was set to 10 rad/s for a total of 10 min. Each cycle lasted 2 min.

Swelling properties, in vitro degradation

Swelling rate: A 5 ml centrifugal tube was filled with 0.5 ml of hydrogel precursor solution. The gelatinized and freeze-dried hydrogel was then incubated by adding equal volumes of PBS. Several times during the incubation period, the weight of each sample was measured, and the percentage swelling of the gel was determined.

Degradation rate: An amount of pig skin hydrogel precursor is placed into a 15 ml centrifugal tube, after gelation, 10 ml PBS solution is added, and the centrifugal tube is placed at 23 °C, 37 °C, and 42 °C. After a certain period to suck out the liquid in the centrifugal tube, weigh the sample.

Release of levofloxacin

In order to investigate the release of Levofloxacin from Lev@PADM hydrogel under varying conditions, standardized UV absorbance curves were subsequently established to calibrate the measurements. The release of Levofloxacin from the Lev@PADM hydrogel was then evaluated at different temperatures. Specifically, 500 μl of Lev@PADM with a drug concentration of 20 μg/ml was placed in centrifuge tubes containing 500 μl of phosphate buffer and incubated at temperatures of 25, 37, or 42 °C. Subsequently, 100 μl of the resulting liquid was analyzed and.added back to the centrifuge tube. Repeat this procedure until the OD of the solution stops increasing.

Cell culture experiment

To perform the cell culture experiment, the Lev@PADM hydrogel was diluted with 1640 medium to obtain concentrations of 10%, 5%, and 2.5% in the overall culture system. These culture systems were then utilized for the cultivation of the 5637 cell line over a period of three consecutive days. The MTT assay kit was employed to assess cell proliferation and growth status during the culture process. Furthermore, the PI/Annexin V assay kit was utilized to detect cell apoptosis.

Continuous antibacterial test

To conduct the continuous antibacterial test, the Lev@PADM hydrogel was diluted with MHB medium, with a gradual decrease in levofloxacin concentration from 2000 ng/ml to 3.90625 ng/ml in a gradient fashion. The resulting medium was then utilized for the cultivation of the CFT073 strain for a duration of 24 h. Throughout the cultivation process, the OD600 value of the CFT073 strain was continuously monitored using a spectrophotometer.

Bacteriostatic circles were measured using a standardized method. Specifically, logarithmic phase cultures of Escherichia coli, Pseudomonas aeruginosa, and Staphylococcus aureus were mixed with MHA medium at a density of 105 CFU/ml. This mixture was then poured into culture plates containing pre-placed Oxford cups. Once the medium solidified, 200 μl of sterile saline, levofloxacin, Lev@PADM hydrogel, and PADM were added to the respective Oxford cups. The plates were subsequently incubated for a period of 7 days. Following the incubation period, the diameters of the bacteriostatic circles were measured.

Mouse models of E.coli-induced bacterial cystitis

Seven-to-eight-week-old Female C57BL/6 mice were purchased from GemPharmatech Co. Ltd. (Nan**g, China). Mice were allowed to recover from the stress of transport in our institutional animal facility for 3 days before the experiments were carried out. All mice were housed in the specific pathogen-free (SPF)-grade individually ventilated cages (IVCs) and maintained under a 12/12 h light/dark cycle at a temperature of 20–26 °C and relative humidity of 40–70%. The mice were provided food and sterilized water ad libitum. All efforts were made to minimize animal suffering. The animal study was approved by the Animal Experimentation Ethics Committee of the Anhui Medical University (approval no. LLSC20230023), and all animals in this study were treated according to the guidelines of Anhui Medical University and with the approval of the Animal Care and Ethics Committee of Animal Usage of Anhui Medical University (Hefei, China).

A modified version of the previously described unobstructed ascending UTI mouse model was employed. To inoculate the mice with bacteria, a bacterial culture was prepared by incubating it overnight at 37 °C in LB broth. The culture was then reinoculated in fresh LB broth and centrifuged at 5000 rpm for 10 min. The resulting pellet was dissolved in 10 ml of sterile PBS. The cell densities were adjusted to 5 × 107 CFUs per 25 μl for the prototypic UPEC strain, CFT073.

After the mice were anesthetized with isoflurane, a syringe covered with a sterile urinary catheter was gently inserted into the urethra. Once the catheter was fully inserted, 50 μl of the bacterial inoculum was slowly injected into the bladder by depressing the syringe plunger. Thirty minutes after urethral inoculation with CFT073, the mice were anesthetized again and, upon regaining their normal mobility, were returned to their original cages.

To ensure the stability of the infection model, the same infection procedure was repeated on the second day. On the third day, the infected mice were subjected to different treatments involving bladder instillation. These treatments included instilling 50 μl of saline, levofloxacin, PADM, and Lev@PADM into the bladder, respectively.

After seven days, the mice were sacrificed. Bladder tissue samples were collected, fixed in 4% (w/v) paraformaldehyde, pre-cooled in liquid nitrogen, and stored at − 80 °C for further processing. Serum and urine samples were also collected and stored at − 80 °C.

Immunohistochemistry and immunofluorescence double staining of bladder tissue

For histopathology analysis, bladder sections were cut into 4-μm-thick slices and rehydrated using a series of ethanol solutions with increasing concentrations. Antigen retrieval was performed by heating the slides in a microwave at 95 °C for 15 min, followed by cooling to room temperature and washing with Tris-buffered saline containing 0.1% (v/v) Tween® 20 Detergent (TBST) wash buffer. To block non-specific protein binding, the slides were incubated with normal goat serum (Gene Tech). Immunohistochemistry was performed using specific primary antibodies, including anti-iNos (18985-1-AP, Proteintech), anti-CD206 (18704-1-AP, Proteintech), anti-MPO (22225-1-AP, Proteintech), and anti-Mast (18189-1-AP, Proteintech). Immunofluorescent staining was performed using anti-E-cadherin (#BF0219, Affinity Biosciences) and anti-E.coli cells (#ab137967; Abcam) antibodies. The slides were incubated with the primary antibodies overnight at 4 °C, followed by washing with TBST. Detection of the target proteins was achieved using either HRP-conjugated secondary antibodies for immunohistochemistry or fluorescent-conjugated secondary antibodies for immunofluorescence staining. Images were captured using a fluorescence microscope (Olympus, Tokyo, Japan) with consistent acquisition parameters.

Detection of levofloxacin concentration

Mice were treated with 50 μl of levofloxacin solution via oral gavage and bladder injection, respectively, while another group of mice was treated with 50 μl of Lev@PADM hydrogel via bladder instillation. All three intervention groups received an equal dose of levofloxacin (20 μg/ml). Urine samples were collected from the mice at 3/6/24/48 h after treatment, and the concentration of levofloxacin in the urine was determined.

The impact of Lev@PADM on intestinal microbiota structure

The mice were divided into three groups: control group, levofloxacin oral gavage group, and Lev@PADM bladder instillation group. Each group received the respective intervention treatment continuously for seven days. Fecal samples were collected from the mice on Day 8 and immediately stored at – 80 °C. Total bacterial genomic DNA was extracted from the fecal samples using the QIAamp 96 Powerfecel QIAcube HT Kit (Qiagen, Germany) according to the manufacturer’s instructions. Amplicon sequencing of the V3-V4 region of the bacterial 16S rRNA gene was performed using primers 343F (5’ TACGGRAGGCAGCAG 3’) and 798R (5’ AGGGTATCTAATCCT 3’) on an Illumina HiSeq 2500 System at Personalbio Biotech Co., Ltd. (Shanghai, China). The raw sequencing data were obtained in FASTQ format. Preprocessing of the paired-end reads was performed using Trimmomatic software to detect and remove ambiguous bases (N). Low-quality sequences (average quality score less than 20) were trimmed using a sliding window approach, followed by read assembly using the FLASH software. QIIME software (version 1.8.0) was used for further denoising of the sequences. Operational taxonomic units (OTUs) were defined using the Vsearch software with a 97% similarity cutoff. Representative reads from each OTU were selected using the QIIME package. All representative reads were annotated by blasting against the Silva database Version 138 using the RDP classifier with a confidence threshold of 70%. Analysis of various parameters such as α-diversity and β-diversity of the microbial community was conducted using the GENESCloud platform developed by Personalbio Biotech Co., Ltd.

Statistical analysis

Data are expressed as mean ± SEM. Statistical significance was evaluated using an unpaired two-tailed Student’s t-test and one-way ANOVA among more than two groups. Differences were considered statistically significant at p < 0.05. GraphPad Prism version 7.0 (GraphPad Software) was used for the statistical analyses.