Effects of Luteolin-7-O-Glucoside on Intestinal Microbiota Dysbiosis and Drug Resistance Transmission Caused by Raoultella ornithinolytica B1645-1: Modulating the Composition of Intestinal Microbiota and Promoting the Transfer of blaNDM-1 Gene from Genus Enterococcus to Lactobacillus in Mice

Raoultella ornithinolytica is an Enterobacteriaceae bacterium that can infect both humans and animals, while luteolin-7-O-glucoside (IOG) is a flavonoid that has broad effects on the intestinal microbiota of healthy animals. However, current studies lack sufficient data on intestinal microbiota dysbiosis and drug resistance transmission caused by R. ornithinolytica and the possible role of IOG. In this study, BALB/c mice were infected with R. ornithinolytica carrying blaNDM-1 gene and treated with IOG (3 mg/kg·d and 6 mg/kg·d) to analyze the diversity of intestinal microbiota and the transfer of blaNDM-1 between bacteria. The findings indicated that R. ornithinolytica B1645-1 exhibited a significant ability to enhance the Firmicutes/Bacteroidota ratio and increase the relative abundance of Lactobacillus and Bacillus after 48 h, where as 6 mg/kg·d IOG had an opposite effect. Moreover, R. ornithinolytica B1645-1 facilitated the emergence of drug-resistant bacteria and promoted blaNDM-1 gene transfer in Enterococcus, Escherichia, Klebsiella, Acinetobacter, Bacillus, Brevibacterium, and Lactobacillus. Enterococcus was the predominant genus at 48 h. Surprisingly, 6 mg/kg·d IOG significantly inhibited the production of drug-resistant bacteria and promoted blaNDM-1 gene transfer from Enterococcus to Lactobacillus at 144 h. However, the role of Lactobacillus as a recipient for drug-resistant genes should be of more concern.

Oral administration is arguably the most effective and easy means of drug delivery that is widely recommended.Following ingestion, IOG is not destroyed by gastric acidand is mainly hydrolyzed by β-glucosidase of gastrointestinal mucosa or intestinal microbiota, such as Enterococci, Lactobacilli, Bacteroides, and Bifidobacteria, to release luteolin.The small intestine absorbs luteolin, which is subsequently re-secreted into the intestine via hepatic excretion [11].Although Enterococcus is particularly active in the metabolism of IOG, it is likely that strictly anaerobic organisms, such as Lactobacilli and Bifidobacteria,are primarily responsible for the hydrolysis of glycoside in the small intestine [11].The metabolism of flavonoids is both influenced by and influences the composition of the intestinal microbiota.Kondapalli NB et al. reported that IOG significantly increased the levels of Lactobacillus and Bifidobacterium in the intestinal microbiota of healthy rodents [12].In vitro studies have shown the antibacterial activity of IOG against Staphylococcus aureus, Streptococcus pneumoniae, Bacillus subtilis, Enterococci, Salmonella typhimurium, and Escherichia coli [13,14].Therefore, it is plausible to suggest that the metabolism and biological activity of IOG may also influence the intestinal microbiota in vivo.
The mammalian gut contains an extremely dense microbial community (>10 12 bacteria/g) that has a significant impact on the host's immune systems [15].Inflammatory responses from the gut immune system or pathogens can lead to suppression of anaerobic microbiota (e.g., Lactobacillus and Bifidobacterium) and boost the colonization density of Enterobacteria, such as Klebsiella pneumoniae and E. coli [16].Many Lactobacillus species and Bifidobacterium species are generally considered safe and are often utilized as probiotics to ameliorate memory deficits, brain neuron damage, glial activation, and fecal microbiota composition [17].Anaerobic bacteria play a crucial role in regulating inflammation, as demonstrated by the attenuation of intestinal inflammation in breast-fed infants supplemented with Bifidobacterium longum subsp.infantis [18].Intestinal inflammation also facilitates horizontal gene transfer among bacterial populations, primarily through persister cells formed by pathogenic bacteria [19], which can promote plasmid transfer up to 99% between different Enterobacteriaceae strains.Enterobacteriaceae have the ability to acquire, accumulate, and disseminate resistance genes via mobile genetic elements from intestinal microbiota [20,21].The persistent emergence of drug-resistant Enterobacteriaceae results in escalating morbidity, mortality, and healthcare costs [22].
Raoultella ornithinolytica is a Gram-negative aerobic bacterium belonging to the Enterobacteriaceae family [23].It is primarily found in aquatic environments, soil, insects, and fish.This bacterium has the ability to convert histidine to histamine, which can cause fish poisoning [24].In humans, it can lead to infections of the digestive tract, urinary tract, and blood, especially in immunocompromised patients [23,25].Although R. ornithinolytica has been found to have weak pathogenicity in some studies, there is clinical concern about the impact of this strain and its multi-drug resistant variants on patients [26].
Current research lacks sufficient data on intestinal microbiota dysbiosis and the transmission of drug resistance caused by R. ornithinolytica.The potential role that IOG can play in regulating these effects is intriguing.To bridge this gap, we established a mouse infection model carrying the bla NDM-1 gene encoding New Delhi Metallo-β-Lactamases-1 (NDM-1), a carbapenemase.The aim was to investigate the impact of R. ornithinolytica on the microbiota in mice and evaluate the regulatory capacity of IOG on intestinal microbiota.
IOG and imipenem (IPM) were purchased from Tianjin Vientiane Hengyuan Technology Co., Ltd.(Tianjin, China) and the National Institute for the Control of Pharmaceutical and Biological Products (Beijing, China), respectively.

Animals and Sample Preparation
BALB/c female mice (6-8 weeks old, 16-18 g) supplied by LCBL (Liaoning Changsheng Biotechnology Limited by Share Ltd., Benxi, China) were housed in cages measuring 25 cm × 18 cm × 13 cm and maintained under controlled temperature conditions of 23 ± 1 • C with a light/dark cycle of 12 h:12 h.They were given a standard rodent animal diet (commercial food pellets) and drinking water ad libitum for one week prior to the experiments, after which they underwent a fasting period of twelve hours.The experiments were conducted at consistent times (08:00 a.m. and 15:00 p.m.) to eliminate any variations due to time of day.
In total, 32 mice were randomly assigned to four groups, with 8 in each group: control group (Con), infection group (IF), infection group treated with 3 mg/kg•d IOG (IFL 3 ), and infection group treated with 6 mg/kg•d IOG (IFL 6 ).
R. ornithinolytica B1645-1 was cultured in Mueller-Hinton liquid broth added with 0.5 × minimal inhibitory concentration (MIC) IPM (MIC > 8 µg/mL) at 37 • C with 200 rpm until the OD 600 of 1.5-2.0.First, all of the IF, IFL 3 , and IFL 6 mice were administrated bacterial solution [350 µL, 1.0×10 10 colony-forming units (CFU)/mL] viagastric intubation for 24 h.Subsequently, the IFL 3 and IFL 6 groups received IOG at doses of 3 mg/kg•d and 6 mg/kg•d (200 µL) for five consecutive days, respectively.The Con was given an equivalent volume of 0.9% normal saline under identical conditions.At both the 24 h and 120 h time points after administration of IOG, the mice were euthanized.Fresh transverse colon contents (approximately 0.1 g) measuring approximately 1 cm in length were dissected and flash-frozen in liquid nitrogen for 30 s before being stored at −80 • C until DNA extraction for microbial diversity and bla NDM-1 gene analysis.
All experiments using bacteria, IPM, IOG, and mice were performed at the Medical Center of Hubei University of Medicine (Licence No.: SYXK 2017-0093).The animal experiments were conducted upon approval from the Ethical Committee for Vertebrate Experiments of Hubei University of Medicine (Ethical approval No. 2020-086).

DNA Extraction, PCR Amplification, and Sequencing
Under ice-water bath conditions, 0.1 g of the sample was immersed in 1 mL of sterile phosphate-buffered saline (PBS) (0.05 M, pH 7.4), vigorously vortexed, and then centrifuged at 200 rpm for 5 min to remove any remaining coarse particles.After this procedure was repeated three times, the combined supernatant was used to precipitate bacteria at 10,000-12,000 rpm for 10 min.
The E.Z.N.A. ® soil DNA kit (Omega) was utilized to extract DNA from the contents of mouse transverse colon.The V3-V4 regions of 16S rDNA were amplified with primers 338F (5 -ACTCCTACGGGAGGCAGCAG-3 ) and 806R (5 -GGACTACHVGGGTWTCTA AT-3 ).The amplification reaction was as follows: 95 • C pre-degeneration for 3 min, 95 • C degeneration for 30 s, 55 • C annealing for 30 s, and 72 • C extension for 30 s, for a total of 27 cycles, and then extension at 72 • C for 10 min.The raw sequence data were submitted to the NCBI's Short Read Archive (SRA) (BioProject ID: PRJNA898665).

Microbial Diversity Analysis
The sequence data were processed using Quantitative Insights Into Microbial Ecology (QIIME2; v2022.2) [28].Paired-end sequences were aligned using the R software (version 4.1) and processed using the Divisive Amplicon Denoising Algorithm 2 (DADA2) version 1.20.0workflow, followed by mapping to amplicon sequence variants (ASVs) for those with >100% pairwise identity [28,29].ASVs were taxonomically classified using QIIME with the SILVA reference database (version 138) to construct a phylogenetic tree through the IQ-TREE [30].Alpha diversity was used to assess species richness within group based on the Chao, ACE, Shannon, and Simpson indices.Beta diversity was performed to measure differences in abundance between groups with non-metric multidimensional scaling (NMDS) at the ASV level [31].Linear discriminant analysis effect size (LEfSe) was used for multi-level species difference discriminant analysis, with a logarithmic linear discriminant analysis (LDA) score > 2 and p < 0.05 considered significant [32].

Screenation and Identification of Strains Carrying Bla NDM-1 Gene
Based on the operation flow shown in Figure 1, 0.1 g of mouse transverse colon content was mixed in 1 mL of sterile saline and centrifuged at 4000 rpm for 2 min.After three repetitions, the combined supernatant was centrifuged at 12,000 rpm for 2 min at 4 • C to enrich bacteria.The bacteria were inoculated into Luria-Bertani liquid medium (10 g/L NaCl, 10 g/L Tryptone, and 5 g/L Yeast Extract) without antibiotics and routinely cultured overnight at 37 • C.Then, 100 µL of the suspension diluted with saline was coated on LB agar medium (10 g/L NaCl, 10 g/L Tryptone, 5 g/L Yeast Extract, and 15 g/L Agar) containing IPM (4 µg/mL) at 37 • C for 48 h to count CFUs.In the same way, anaerobic agar medium (Pancreatic Casein Peptone 20 g/L, Sodium Chloride 5 g/L, Sodium Formaldehyde Sulfite 1 g/L, Water-Soluble Aniline Blue 0.002 g/L, Glucose 10 g/L, Agar 20 g/L, and Sodium Thioglycolate 2 g/L) with 4 µg/mL IPM was used to separate anaerobic bacteria from the samples [33].CFUs from plates containing IPM were considered as bla NDM-1 -positive colonies.
(ASVs) for those with >100% pairwise identity [28,29].ASVs were taxonomically classified using QIIME with the SILVA reference database (version 138) to construct a phylogenetic tree through the IQ-TREE [30].Alpha diversity was used to assess species richness within group based on theChao, ACE, Shannon, and Simpson indices.Beta diversity was performed to measure differences in abundance between groups with non-metric multidimensional scaling (NMDS) at the ASV level [31].Linear discriminant analysis effect size (LEfSe) was used for multi-level species difference discriminant analysis, with a logarithmic linear discriminant analysis (LDA) score > 2 and p < 0.05 considered significant [32].

Screenation and Identification of Strains Carrying BlaNDM-1 Gene
Based on the operation flow shown in Figure 1, 0.1 g of mouse transverse colon content was mixed in 1 mL of sterile saline and centrifuged at 4000 rpm for 2 min.After three repetitions, the combined supernatant was centrifuged at 12,000 rpm for 2 min at 4 °C to enrich bacteria.The bacteria were inoculated into Luria-Bertani liquid medium (10 g/L NaCl, 10 g/L Tryptone, and 5 g/L Yeast Extract) without antibiotics and routinely cultured overnight at 37 °C.Then, 100 µL of the suspension diluted with saline was coated on LB agar medium (10 g/L NaCl, 10 g/L Tryptone, 5 g/L Yeast Extract, and 15 g/L Agar) containing IPM (4 µg/mL) at 37 °C for 48 h to count CFUs.In the same way, anaerobic agar medium (Pancreatic Casein Peptone 20 g/L, Sodium Chloride 5 g/L, Sodium Formaldehyde Sulfite 1 g/L, Water-Soluble Aniline Blue 0.002 g/L, Glucose 10 g/L, Agar 20 g/L, and Sodium Thioglycolate 2 g/L) with 4 µg/mL IPM wasused to separate anaerobic bacteria from the samples [33].CFUs from plates containing IPM were considered as blaNDM-1-positive colonies.To determine the accuracy of blaNDM-1-positive-colony screening results, PCR was used to detect the blaNDM-1 gene.Amplification was conducted with the primer set F(5′-CTCGCACCGAATGTCTGGC-3′) and R(5′-GGGGCGTAGTGCTCAGTGTC-3′).The PCR amplification reactions included 25 µL of 2 × NovoStar Green PCR Mix, 2 µL each of upstream and downstream primers (1 mM), 1 µL of DNA template, and ddH2O added to a total volume of 50 µL.The PCR conditions were as follows: pre-denaturation at 95 °C for 5 min, denaturation at 95 °C for 30 s, annealing at 55 °C for 30 s, and extension at 72 °C for 1 min, for a total of 34 cycles, and then extension at 72 °C for 5 min.The amplified products were separated via1% agarose gel electrophoresis (110 V, 30 min) and sequenced.The gene sequence was compared with the GenBank database to determine the blaNDM-1 gene, and the bacterial strains carrying blaNDM-1 were identified usingmass spectrometry method [34].To determine the accuracy of bla NDM-1 -positive-colony screening results, PCR was used to detect the bla NDM-1 gene.Amplification was conducted with the primer set F(5 -CTCGCACCGAATGTCTGGC-3 ) and R(5 -GGGGCGTAGTGCTCAGTGTC-3 ).The PCR amplification reactions included 25 µL of 2 × NovoStar Green PCR Mix, 2 µL each of upstream and downstream primers (1 mM), 1 µL of DNA template, and ddH 2 O added to a total volume of 50 µL.The PCR conditions were as follows: pre-denaturation at 95 • C for 5 min, denaturation at 95 • C for 30 s, annealing at 55 • C for 30 s, and extension at 72 • C for 1 min, for a total of 34 cycles, and then extension at 72 • C for 5 min.The amplified products were separated via 1% agarose gel electrophoresis (110 V, 30 min) and sequenced.The gene sequence was compared with the GenBank database to determine the bla NDM-1 gene, and the bacterial strains carrying bla NDM-1 were identified using mass spectrometry method [34].

Statistical Analysis
One-way ANOVA with Tukey's multiple comparison test and Kruskal-Wallis H test were used to compare the data among groups, and p < 0.05 was considered to be statistically significant [35].MetaPhlAn2 was used to determine species abundances [36].GraphPad Prism v8.0 and R package were used to generate graphs for this study.

The Diversity of Intestinal Microbiota
A total of 3,829,368 sequences (>202 bp) were derived from 32 samples.After DADA2 noise reduction, 1,541,391 effective sequences were retained and 10,585 ASV sequences were clustered according to the minimum number of sample sequences (16,326 sequences) (due to the death of one mouse and the lack of sequences for two mice, three inefficient samples were eliminated) (Table S1).The rarefaction curves of various samples reached a plateau, indicating that the sampling was effective and the ASV datasets were successfully recovered (Figure S1).
The Chao and ACE indices were used to determine species richness, while the Simpson and Shannon indices were used to estimate bacterial diversity (Figure 2).At 48 h, significant differences in species richness estimated based on the Chao and ACE indices were observed among the four groups, as demonstrated by Figure 2A,B.After infection, the species richness in the IF group was significantly lower than the Con group (p < 0.05) (Figure 2A), and the species richness was not recovered until the dosage of IOG reached 6 mg/kg•d (IFL 6 ) (Figure 2B).The analysis of bacterial diversity estimated based on the Simpson and Shannon indices revealed variations among the different life stages, although these differences did not reach statistical significance (Figure 2C,D).No significant difference in species richness (Figure 2E,F) or diversity (Figure 2G,H) was observed among the four groups at 144 h (p > 0.05).

Statistical Analysis
One-way ANOVA with Tukey's multiple comparison test and Kruskal-Wallis H test were used to compare the data among groups, and p < 0.05 was considered to be statistically significant [35].MetaPhlAn2 was used to determine species abundances [36].GraphPad Prism v8.0 and R package were used to generate graphs for this study.

The Diversity of Intestinal Microbiota
A total of 3,829,368 sequences (>202 bp) were derived from 32 samples.After DA-DA2 noise reduction, 1,541,391 effective sequences were retained and 10,585 ASV sequences were clustered according to the minimum number of sample sequences (16,326 sequences) (due to the death of one mouse and the lack of sequences for two mice, three inefficient samples were eliminated) (Table S1).The rarefaction curves of various samples reached a plateau, indicating that the sampling was effective and the ASV datasets were successfully recovered (Figure S1).
The Chao and ACE indices were used to determine species richness, while the Simpson and Shannon indices were used to estimate bacterial diversity (Figure 2).At 48 h, significant differences in species richness estimated based on the Chao and ACE indices were observed among the four groups, as demonstrated by Figure 2A,B.After infection, the species richness in the IF group was significantly lower than the Con group (p < 0.05) (Figure 2A), and the species richness was not recovered until the dosage of IOG reached 6 mg/kg•d (IFL6) (Figure 2B).The analysis of bacterial diversity estimated based on theSimpson and Shannon indices revealed variations among the different life stages, although these differences did not reach statistical significance (Figure 2C,D).No significant difference in species richness (Figure 2E,F) or diversity (Figure 2G,H) was observed among the four groups at 144 h (p > 0.05).Figure 3 shows that NMDS had a better fitting degree in ASV levels at 48 h and 144 h (stress = 0.055).The difference between groups at 48 h was greater than the difference within groups (R = 0.3577, p = 0.009).There were significant differences in community composition between groups (p < 0.05), indicating that all four groups were effective Figure 3 shows that NMDS had a better fitting degree in ASV levels at 48 h and 144 h (stress = 0.055).The difference between groups at 48 h was greater than the difference within groups (R = 0.3577, p = 0.009).There were significant differences in community composition between groups (p < 0.05), indicating that all four groups were effective (Figure 3A).However, at 144 h, there was no significant difference observed between the Con, IF, IFL 3 , and IFL 6 groups (p > 0.05), indicating that IOG had no further impact on the community (Figure 3B).(Figure 3A).However, at 144 h, there was no significant difference observed between the Con, IF, IFL3, and IFL6 groups (p > 0.05), indicating that IOG had no further impact on the community (Figure 3B).

Taxonomic Comparison of Intestinal Microbiota
The compositions of the intestinal microbiota at phylum and genus levels were characterized.The community barplot analysis showed that the microbiota of 29 samples mainly covered seven phyla, including Firmicutes, Bacteroidota, Campilobacterota, Desulfobacterota, Patescibacteria, Actinobacteriota, and Deferribacterota (Figure 4A,B), of which Firmicutes and Bacteroidota were the dominant phyla.

Taxonomic Comparison of Intestinal Microbiota
The compositions of the intestinal microbiota at the phylum and genus levels were characterized.The community barplot analysis showed that the microbiota of 29 samples mainly covered seven phyla, including Firmicutes, Bacteroidota, Campilobacterota, Desulfobacterota, Patescibacteria, Actinobacteriota, and Deferribacterota (Figure 4A,B), of which Firmicutes and Bacteroidota were the dominant phyla.
Norank_f__Muribaculaceae and Lactobacillus were clearly the dominant genera in the Con group (25.46% and 16.68%).At 48 h (Figure 5A), the IF group showed a significant increase of 59.23% in the relative abundance of Lactobacillus (p < 0.05).Although there was a decreasing trend in the relative abundance of norank_f__Muribaculaceae in the IF group, this difference did not reach statistical significance (p > 0.05).Following the administration of IOG, the relative abundance of Lactobacillus significantly decreased from 55.72% in the IFL 3 group to 31.13% in the IFL 6 group, and when the dosage was 6 mg/kg•d (IFL 6 ), the relative abundance of Lactobacillus recovered to a level with no significant difference compared to the Con group (16.68%).This result suggested that IOG possesses the po-tential to reduce the increase in Lactobacillus count due to infection with R. ornithinolytica B1645-1.The relative abundance of Lactobacillus, Helicobacter, unclassified_f__Rikenellaceae, Candidatus_Saccharimonas, norank_f__Lachnospiraceae, and Skermanella exhibited significant differences between groups (p < 0.05) (Figure 5C).The LEfSe analysis showed that Lactobacillus species had the greatest effect in the IF, IFL 3 , and IFL 6 groups (Figure 6A).Norank_f__MuribaculaceaeandLactobacillus were clearly the dominant genera in t Con group (25.46% and 16.68%).At 48 h (Figure 5A), the IF group showed a significa increase of 59.23% in the relative abundance of Lactobacillus (p< 0.05).Although there w  Figure 6.Linear discriminant analysis effect size (LEfSe) at the genus level in mice between Con, IFL3, and IFL6 groups at 48 h (A) and 144 h (B).LEfSe, linear discriminant analysis effect size; LE scores > 2 are shown.The prefixes for taxonomic ranks are represented as follows: "p" for phylu "c" for class, "o" for order, "f" for family, and "g" for genus.
As shown in Figure 5B (at 144 h), the relative abundance of unclassi-fied_f__Lachnospiraceae in the IF, IFL3, and IFL6 groups increased gradually, while that of Lactobacillus significantly decreased to 7.93%, 1.7%, and 5.66%, respectively (15.79% in Con). Figure 5D shows that Prevotellaceae_UCG-001, Eubacterium_siraeum_group, Streptococcus, and Lachnospiraceae_UCG-004 were significantly different species (p < 0.05).The LEfSe analysis also showed that Prevotellaceae_UCG-001species and Streptococcus species had the greatest impact in the IFL3 and IFL6 groups, respectively (Figure 6B). Figure 6.Linear discriminant analysis effect size (LEfSe) at the genus level in mice between Con, IF, IFL 3 , and IFL 6 groups at 48 h (A) and 144 h (B).LEfSe, linear discriminant analysis effect size; LEfSe scores > 2 are shown.The prefixes for taxonomic ranks are represented as follows: "p" for phylum, "c" for class, "o" for order, "f" for family, and "g" for genus.
In conclusion, with the extension of infection time, the number of aerobic and anaerobic resistant bacteria significantly increased.However, the number in the IFL 6 group remained consistently lower than that in the IF and IFL 3 groups, indicating IOG's inhibitory effect on the production of drug-resistant bacteria.

Identification of Strains Carrying Bla NDM-1 Gene
At 48 h (Table 1), seven genera and eight species of bla NDM-1 -positive bacteria were successfully isolated from the IF, IFL 3 , and IFL 6 groups, including Lactobacillus johnsonii, Enterococcus faecalis, Enterococcus gallinarum, Klebsiella pneumoniae, Escherichia coli, Brevibacterium linens, Acinetobacter baumannii,and Bacillus pumilus, of which Enterococcus was the dominant genus.However, at 144 h, there were still bla NDM-1 -positive bacteria in five genera (Enterococcus, Chromobacterium, Lactobacillus, Microbacterium, and Escherichia) in the IF, IFL 3 , and IFL 6 groups (Table 1).Except for the Con group, Lactobacillus was the dominant genus in all groups, with a total of four species, including L. johnsonii, L. reuteri, L. murinus, and L. gasseri.
In conclusion, with the extension of infection time, the number of aerobic and anaerobic resistant bacteria significantly increased.However, the number in the IFL6 group remained consistently lower than that in the IF and IFL3 groups, indicating IOG's inhibitory effect on the production of drug-resistant bacteria.

Discussion
Enteropathogenic infections and infection-mediated inflammatory responses can disrupt the intestinal ecosystem, including changing the composition and number of intestinal microbiota [16].R. ornithinolytica B1645-1 isolated from the blood sample of a renal transplant patient who died of sepsis was found to be pathogenic and significantly disrupted the composition of intestinal microbiota in mice.Specifically, the Firmicutes/Bacteroidota ratio was significantly higher in the IF group (5.52:1) than in the Con group (1.1:1).Additionally, at 48 h, there was a notable increase in Lactobacillus abundance from 17% in the Con group to 60% in the IF group.During the initial phase (48 h), total colonic bacterial numbers (ASVs) decreased significantly while Lactobacillus increased, which may be attributed to pathogenic bacterial colonization and subsequent elimination of Bacteroidota [16].However, the Firmicutes/Bacteroidota ratio recovered to 0.68:1 at 144 h and Lactobacillus was no longer the factor that affected the difference (Figure 6B), indicating that colonizing the intestine by R. ornithinolytica alone was not enough to continuously reduce the overall level of intestinal microbiota and alter the proportion of resident bacteria.The effect of R. ornithinolytica B1645-1 on the intestinal microbiota of mice was also evident in alterations to other bacteria, such as norank_f__Muribaculaceae and Bacillus (Figure 5A,B, Table S2).
The major groups of mammalian intestinal microbiota include Lactobacilli, Bifidobacteria, Enterobacteria, Enterococci, and Bacteroides species [11].In animals, Firmicutes and Bacteroidota (such as Bacteroides and Bifidobacteria), which are strictly anaerobic bacteria, maintain a relatively low level of facultative anaerobes (such as Enterobacteriaceae and Enterococci), while Lactobacilli are abundant in rats and mice [11,15].IOG treatment for 24 h significantly altered the composition of intestinal microbiota in all groups.The Firmicutes/Bacteroidota ratio was significantly lower in the IFL 6 group (2.91:1) compared to the IFL 3 (7.46:1)and IF (5.52:1) groups, indicating that 6 mg/kg•d IOG had a stronger regulatory effect on intestinal microbiota recovery in infected mice than in other groups.Contrary to previous reports suggesting that IOG promotes an increase in Lactobacillus within the intestinal flora of healthy rodents [12], our study surprisingly found a decrease in Lactobacillus numbers from 50% in the IFL 3 group to 30% in the IFL 6 group at 48 h.This regulatory effect was even more significant at 144 h, indicating that IOG primarily targets Lactobacillus as one of the main influencers on intestinal microbiota.
The spread of carbapenemase (KPC, NDM, OXA-48, and VIM)-producing enterobacteria (CPE) clones among patients and the transfer of carbapenemase-encoding genes between enterobacteria in individual patients' intestinal microbiota have shaped the epidemiological characteristics of CPE [23,37].However, the transfer of carbapenemase-encoding genes within the intestinal microbiota of individual patients remains poorly understood.The ability of IOG to modulate the intestinal microbiota has been demonstrated through alterations in phyla (Bacteroidota and Firmicutes) and genera (Lactobacillus, Bacillus, etc.), which are associated with the transmission of drug-resistant genes within intestinal microbiota.After 48 h of infection, the bla NDM-1 gene carried by R. ornithinolytica B1645-1 was detected in various intestinal bacteria, including E. faecalis, E. gallinarum, E. coli, K. pneumoniae, A. baumannii, B. pumilus, B.linens, and Lactobacillus johnsonii, with Enterococcus being the predominant genus.However, after a 120 h treatment with IOG, intestinal bacteria carrying the bla NDM-1 gene underwent significant changes and Lactobacillus (L.johnsonii, L. reuteri, L. murinus, and L. gasseri) became the dominant carrier of the bla NDM-1 gene.This finding indicated that the bla NDM-1 gene had been transferred from pathogenic bacteria (R. ornithinolytica B1645-1) to Enterococcus within 48 h, and subsequently to Lactobacillus within 144 h, either through natural recovery in the IF group or after treatment with IOG in the IFL 3 and IFL 6 groups.Although IOG exhibited efficacy in reducing the number of bacterial genera or species carrying drug-resistant genes and the abundance of drug-resistant bacteria (Figure 7 and Table 1), Lactobacillus, as a probiotic with intestinal advantages [38], has become the carrier of drug-resistant genes and is at risk of expanding the transmission of drug-resistant genes.
R. ornithinolytica B1645-1 induced a shift in the microbial composition, leading to ecological disorder.This finding has significant implications for comprehending the biological function of R. ornithinolytica and for clinical prevention and treatment.IOG rectified this interference and facilitated the translocation of the bla NDM-1 gene from Enterococcus to Lactobacillus, providing strong support for its clinical application.However, as a recipient for resistant genes, Lactobacillus warrants further investigation.

Conclusions
R. ornithinolytica administered orally is sufficient to perturb the composition of intestinal microbiota and facilitate the transfer of the bla NDM-1 gene in intestinal microbiota.Luteolin-7-O-glucoside can modulate this disordered intestinal microbiota, inhibit the production of drug-resistant bacteria, and promote the transfer of the bla NDM-1 gene from the genus Enterococcus to Lactobacillus.However, Lactobacillus should be paid more attention as the recipient of drug-resistant genes because it may be an inevitable donor.

Figure 2 .
Figure 2. The indices of Alpha diversity (Chao, ACE, Shannon, and Simpson) within four groups at 48 h and 144 h.(A,E) Chao and (B,F) ACE indices were used to determine species richness; (C,G) Shannon and (D,H) Simpson indices were used to estimate bacterial diversity.Significant differences are denoted by asterisks according to the t-test (*: p < 0.05; **: p < 0.01).

Figure 2 .
Figure 2. The indices of Alpha diversity (Chao, ACE, Shannon, and Simpson) within four groups at 48 h and 144 h.(A,E) Chao and (B,F) ACE indices were used to determine species richness; (C,G) Shannon and (D,H) Simpson indices were used to estimate bacterial diversity.Significant differences are denoted by asterisks according to the t-test (*: p < 0.05; **: p < 0.01).

Figure 3 .
Figure 3.The indices of Beta diversity between four groups at 48 h and 144 h.The non-metric multidimensional scaling (NMDS) analyses on amplicon sequence variant (ASV) level at 48h (A) and 144h (B) are shown.The stress value represents the fitting degree of NMDS analysis (stress < 0.05, excellent fitting; stress < 0.1, good fitting; stress < 0.2, average fitting; stress > 0.3, poor fitting).Analysis of similarities (ANOSIM) (R: −1−1, R value close to 1 indicates agreater difference between groups) was used to test the difference between groups.

Figure 3 .
Figure 3.The indices of Beta diversity between four groups at 48 h and 144 h.The non-metric multidimensional scaling (NMDS) analyses on amplicon sequence variant (ASV) level at 48 h (A) and 144 h (B) are shown.The stress value represents the fitting degree of NMDS analysis (stress < 0.05, excellent fitting; stress < 0.1, good fitting; stress < 0.2, average fitting; stress > 0.3, poor fitting).Analysis of similarities (ANOSIM) (R: −1−1, R value close to 1 indicates agreater difference between groups) was used to test the difference between groups.

Figure 4 .
Figure 4.The relative abundance of bacterial phyla in the intestinal microbiota of mice.Only ta with a relative abundance ≥ 1% in at least one sample were analyzed at 48 h (A) and 144 h (B).(C represents the Firmicutes/Bacteroidota ratio at 48 h and 144 h, respectively.(E) Phyla with sign cant differences at 48 h.(F) Phyla with significant differences at 144 h.The seven main phyla of testinal microbiota include Firmicutes (green), Bacteroidota (red), Campilobacterota (blue), Desulfob terota (pink), Patescibacteria (orange), Actinobacteriota (yellow), Deferribacterota (purple), and oth (species abundance < 0.01) (brown).The phylum is represented by a bar graph, and the absci represents the ASV level of the phylum.

Figure 4 .
Figure 4.The relative abundance of bacterial phyla in the intestinal microbiota of mice.Only taxa with a relative abundance ≥ 1% in at least one sample were analyzed at 48 h (A) and 144 h (B).(C,D) represents the Firmicutes/Bacteroidota ratio at 48 h and 144 h, respectively.(E) Phyla with significant differences at 48 h.(F) Phyla with significant differences at 144 h.The seven main phyla of intestinal microbiota include Firmicutes (green), Bacteroidota (red), Campilobacterota (blue), Desulfobacterota (pink), Patescibacteria (orange), Actinobacteriota (yellow), Deferribacterota (purple), and others (species abundance < 0.01) (brown).The phylum is represented by a bar graph, and the abscissa represents the ASV level of the phylum.

Figure 5 .
Figure 5.The relative abundance of the bacterial genera in the intestinal microbiota of mice.O taxa with a relative abundance ≥1% in at least one sample were analyzed at 48 h (A,C) and 14 (B,D).

Figure 5 .
Figure 5.The relative abundance of the bacterial genera in the intestinal microbiota of mice.Only taxa with a relative abundance ≥1% in at least one sample were analyzed at 48 h (A,C) and 144 h (B,D).

Figure 5 .
Figure 5.The relative abundance of the bacterial genera in the intestinal microbiota of mice.Only taxa with a relative abundance ≥1% in at least one sample were analyzed at 48 h (A,C) and 144 h (B,D).

Figure 6
Figure6.Linear discriminant analysis effect size (LEfSe) at the genus level in mice between Con, IF, IFL3, and IFL6 groups at 48 h (A) and 144 h (B).LEfSe, linear discriminant analysis effect size; LEfSe scores > 2 are shown.The prefixes for taxonomic ranks are represented as follows: "p" for phylum, "c" for class, "o" for order, "f" for family, and "g" for genus.