A 73-year-old male patient with frequent urination after bladder cancer surgery was admitted to our hospital on 4 July 2022. The patient underwent bladder tumor resection and bilateral ureteral stent placement for bladder tumor in the trigone 40 days ago, and postoperative pathology showed high-grade non-invasive urothelial carcinoma. The laboratory tests revealed increased neutrophils and fasting blood glucose, but decreased erythrocyte, hemoglobin, and albumin. Urinalysis revealed a WBC count of 523/μl (normal 0–6/μl), an erythrocyte count of 9/μl, urine protein of 1+, and urine glucose of 2+. After isolation of the purified bacterial strain, the patient was diagnosed with urinary tract infection. The details of the case report are presented in the Supplementary Material .

After admission, the clean-catch urine of the patient was collected prior to antibiotic therapy and plated onto Columbia blood agar and MacConkey agar plates (Guangzhou Dijing Microbial Technology Co., Ltd., Guangzhou, China) at 35°C in the presence of 5% CO₂. After incubation overnight, bacterial colonies with homogeneous morphology were observed and counted on both agar plates, and the bacterial titer was higher than 10⁵ colony-forming units per milliliter (CFUs/ml).

The isolated strain UI705 was identified on matrix-assisted laser desorption ionization/time of flight mass spectrometry (MALDI-TOF MS, Bruker Daltonik GmbH, Germany) platform by direct smear of fresh bacterial colony on the target plate in accordance with the manufacturer’s instructions. The acquired spectra data from tested bacteria were compared with the existing known spectrum by using the MALDI Biotyper 3.1 software (Bruker Daltonik GmbH, Germany).

All the reagents used for biochemical characterization of strain UI705 were purchased from Hangzhou Microbial Reagent, Co., Ltd (Hangzhou, China). The catalytic activity of catalase was detected using 3% (v/v) H₂O₂. The oxidase activity was measured by filter paper soaked with the reagent tetramethyl-p-phenylenediamine.

Enzymatic activity and acid production from various carbohydrates of the strain were determined by bacterial microbiochemical reaction tubes according to the manufacturer’s instructions. Antimicrobial susceptibility test (AST) was performed by the Kirby-Bauer disc diffusion method with antibiotics (Thermo Fisher Scientific, Waltham, MA, USA).

The genomic DNA of purified bacteria was extracted using the TIANamp Bacteria DNA Kit (TIANGEN Biotech, Co., Ltd, Beijing, China) according to the instructions of the manufacturer. 16S rRNA gene sequencing was performed with universal primers (27F: 5′;-AGTTTGATCMTGGCTCAG-3′;; 1492R: 5′;-GGTTACCTTGTTACGACTT-3′;). The amplification was performed on C1000 Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA) with the following procedure: denaturation at 95°C for 3 min, 30 cycles of denaturation at 95°C for 30 s, annealing at 58°C for 30 s and elongation at 72°C for 1 min, and a further 10 min of elongation at 72°C. The complete 16S rRNA sequence of the strain UI705 was analyzed with the EzBioCloud Database (Yoon et al., 2017a). A phylogenetic tree was constructed using the neighbor-joining method by MEGA software version 11 (Tamura et al., 2021).

The draft genome sequencing of the strain UI705 was performed using the Illumina HiSeq platform by generating paired-end libraries (PE 150 bp). Fragmented genomic DNA with an average size of 300 bp was selected for sequencing. The raw data of sequencing were evaluated by FastQC v0.11.2 and cut by Trimmomatic v0.36 (Bolger et al., 2014) to obtain relatively accurate and effective data. The filtered reads were assembled into contigs using SPAdes v3.5.0 (Bankevich et al., 2012), and GapFiller v1.11 (Boetzer and Pirovano, 2012) is used to fill the gap between contigs. The genetic elements were predicted by Prokka v1.10 (Seemann, 2014). The average nucleotide identity was calculated using the OrthoANIu algorithm (Yoon et al., 2017b). The DNA–DNA hybridization (DDH) value between UI705 and its related type strain was analyzed by Genome-to-Genome Distance Calculator 3.0 (Meier-Kolthoff et al., 2022). The prediction of antibiotic resistance, virulence factors, determinants of pathogen–host interaction, and carbohydrate-active enzymes was performed by means of the Comprehensive Antibiotic Resistance Database (CARD) (McArthur et al., 2013), the Virulence Factors of Pathogenic Bacteria (VFDB) (Chen et al., 2016), the Pathogen–Host Interactions (PHI) database (Urban et al., 2015), and the Carbohydrate-Active enZYmes (CAZy) Database (Lombard et al., 2014), respectively. The automatic annotation server of protein sequence such as NCBI non-redundant protein sequences (NR) (https://www.ncbi.nlm.nih.gov/), Protein family (PFAM) (Finn et al., 2016), Swiss-Prot (UniProt-Consortium, 2015), Clusters of Orthologous Groups of proteins (COG) (Tatusov et al., 2000), Conserved Domain Database (CDD) (Marchler-Bauer et al., 2013), and Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa and Goto, 2000) was applied under the right conditions.

The urine sample developed bacterial colonies on Columbia blood agar after being cultured overnight. The colonies appeared beige to yellow, smooth and circular without apparent hemolytic zones and had diameters between 1 and 2 mm ( Figure 1A ). A spectrogram with protein molecular mass of the strain was acquired by MALDI-TOF MS ( Figure 1B ) and compared with that of known bacterial species in the database provided by the manufacturer. According to the spectra comparison, the strain UI705 was matched to P. anthophila DSM 23080^T with a high confidence level ( Figure 1C ). Biochemical tests revealed that the strain UI705 is catalase-positive and oxidase-negative. β-galactosidase is produced. Indole, H₂S, arginine dihydrolase, and urease are not produced. Acid is produced from glucose, mannose, arabinose, rhamnose, and xylose ( Table 1 ). The AST via the Kirby–Bauer method showed that the strain is susceptible to almost all tested antibiotics except for ampicillin and cefazolin, which belong to penicillin and first-generation cephalosporin antibiotics, respectively ( Table 2 ).

To further investigate the phylogenetic features of the isolated strain UI705 in this study, 16S rRNA gene sequencing was performed. The complete 16S rRNA sequence of the strain UI705 was analyzed with the EzBioCloud Database (Yoon et al., 2017a). The strain UI705 exhibited highest (99.64%) 16S rRNA gene sequence similarity with the type strain of P. anthophila LMG 2558^T (GenBank accession no. EF688010). A total of 1,387 contiguous nucleotides of the strain UI705 were determined and compared with that of P. anthophila LMG 2558^T with a completeness of 98.5%. Among the 1,387 bases, there was only a five-base difference from strain LMG 2558^T. The 16S rRNA sequencing results were submitted to GenBank (accession no. OQ704296). Multiple alignments with sequences of the related Pantoea and the calculations of the levels of sequence similarity were carried out using the MUSCLE algorithm (Edgar, 2004). A phylogenetic tree was constructed using the neighbor-joining method by MEGA software version 11 (Tamura et al., 2021). The topology of the phylogenetic tree was evaluated by using the bootstrap resampling method with 1,000 replicates. The phylogenetic tree showed that strain UI705 was clustered with the type strain LMG 2558^T, and this cluster was strongly supported with a bootstrap value of 82% ( Figure 2 ). The results of the comparative 16S rRNA gene sequence analysis demonstrated that the isolated strain UI705 belongs to the P. anthophila species.

The draft genome sequence of P. anthophila UI705 revealed 4,542,010 bp in size and assembled into 15 contigs, with an N50 value of 564,400 bp and 243.23 × coverage. The GC content of the assembled genome is 56.95% ( Figure 3 ), which is consistent with that of other reports available. Automatic annotation with the NCBI PGAP revealed 4,288 protein coding genes. The details of gene element prediction were listed in Table 3 . Genome-based taxonomy classification using OrthoANIu algorithm validated that UI705 belongs to P. anthophila as it revealed a high ANI value of 99.88% between UI705 and the P. anthophila type strain LMG 2558^T. Moreover, the DDH value between UI705 and its related type strain was analyzed by GGDC 3.0 with an estimated value of 99.30% (Formula 2). The gene function annotation of the UI705 was performed in a database that is based on NCBI Blast+ and KEGG Automatic Annotation Server, including CDD, COG, NR, and PFAM, and the number of annotated genes is the highest in the NR database ( Supplement Figure S1 ). The assembled sequences of P. anthophila UI705 were subjected to GO for further annotation and description of the properties of genes and gene products that were divided into molecular function, cellular component, and biological process ( Figure 4A ). The result from KEGG revealed 100 predictive genes involved in human diseases, mainly those related to infectious diseases and drug resistance ( Figure 4B ), indicating that P. anthophila may emerge as a human pathogen.

The antibiotic resistance of the isolate U1705 was predicted by the CARD and 81 proteins were predicted, such as those related to antibiotic efflux, antibiotic inactivation, and reduced permeability to antibiotic ( Figure 5A ). The antibiotic resistance ontology adeF, rsmA, and CRP were annotated with high percentage identity of matching region (61.44%–89.66%), and the length ratio of reference sequence ranges from 99.24% to 100%. These AROs belong to the resistance-nodulation-cell division (RND) antibiotic efflux pump family and mediate antibiotic resistance in plenty of Gram-negative bacteria (Blair and Piddock, 2009). In addition, a total of 126 virulence factor terms were predicted through the core dataset of the VFDB, which contains 368 predicted proteins ( Figure 5B ). The database predicted the most immune modulation-related proteins, followed by an effector delivery system that includes Type III and VI Secretion Systems (T3SS and T6SS). The T3SS is a needle-like protein complex found in several species of plant or human pathogenic organisms within genus Pantoea, contributing to the pathogenicity to their hosts (Correa et al., 2012; Kirzinger et al., 2015). Based on PHI database, the U1705 exhibited pathogenicity to both animals and plants, with a total of 198 PHI-related proteins predicted ( Figure 5C ). There are 11 annotated genes that are related to reduced virulence or loss of pathogenicity in human host. The carbohydrate active enzyme annotation was carried out by CAZy with e-value < 1e−5 and glycosyl transferases remain the highest ( Figure 5D ). The lipopolysaccharide core heptosyltransferase RfaQ and RfaG, and capsular polysaccharide synthesis protein that belong to glycosyl transferases play important roles in the pathogenicity of many bacteria (Aguiniga et al., 2016; St John et al., 2023).