Characterization of the novel class A β-Lactamase gene bla_KOB−1 in an Enterobacter isolate

Abstract

Background

Class A serine β-lactamases are important enzymes that contribute to bacterial resistance to β-lactam antibiotics, especially in gram-negative bacilli (e.g., Enterobacteriaceae). These enzymes are highly diverse with many molecular variants, and some enzymes (such as SHV and TEM) can be used to extend the inactivation spectrum to the latest generation of cephalosporins by mutation. Consequently, the mechanisms of resistance to antimicrobial agents in bacteria and the clinical implications of antimicrobial resistance need to be more comprehensively studied.

Methods

Animal fecal samples were collected with sterile swabs, and individual colonies were isolated via the agar plate streaking method. The minimum inhibitory concentration (MIC) of the antimicrobial agent was tested by the standard agar dilution method. The complete bacterial genome sequence was obtained using the Illumina and PacBio sequencing platforms. Potential resistance genes were annotated according to the Comprehensive Antibiotic Resistance Database (CARD) and validated by molecular cloning. The β-lactamase KOB-1 was acquired by using the expression vector pCold I, after which the enzymatic kinetics were determined. The evolutionary relationships and genetic environments of the sequences associated with this resistance gene were analyzed via bioinformatics methods.

Results

A novel chromosomal class A β-lactamase gene, bla_KOB−1, was identified in the Enterobacter kobei strain DW108, which was isolated from goose feces. bla_KOB−1 induced varying degrees of resistance to penicillins. Enzyme kinetic parameter assays revealed that KOB-1 could hydrolyze penicillins as well as some cephalosporins (such as cefazolin and cefotaxime), while its hydrolytic activities could be inhibited by tazobactam and avibactam. The bla_KOB−1(-like) genes were uniquely distributed in strains of the genus Enterobacter. Comparative analysis of the bla_KOB−1(-like) gene related sequences showed that, except for a fragment encoding two genes (bla-phoX) that was present only in the sequences from Enterobacter, the gene context of the sequences from the genus Leclercia were similar with that from Enterobacter in terms of gene content or gene order. No mobile genetic element was identified next to the bla_KOB−1 gene.

Conclusion

The identification of additional resistance genes aids in the elucidation of bacterial resistance mechanisms and application of effective clinical treatments for infections caused by pathogens carrying similar resistance genes.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12866-025-04647-4.

Introduction

The Enterobacter cloacae complex (ECC) comprises at least five different species of the genus Enterobacter, including E. cloacae, E. asburiae, E. kobei, E. ludwigii, and E. hormaechei, with different subspecies [1]. Enterobacter species inhabit the guts of animals and humans naturally as part of the commensal microbiota, and only certain subspecies/species can cause infectious diseases. ECC has emerged as an important group of nosocomial pathogens in the last decade that is responsible for nearly 70% of bacteremia caused by Enterobacter species worldwide [2] and is the third most common cause of infections of intra-abdominal foci, the urinary tract, the bloodstream, wounds and so on [3, 4]. E. kobei has been reported to cause nosocomial urosepsis and neonatal sepsis [5, 6]. The majority of E. kobei strains were isolated from clinical specimens, such as urine from catheters and ear discharge, especially intra-abdominal specimens [2, 7].

Enterobacteriaceae bacteria are considered the main cause of resistant nosocomial infections [8] and they naturally harbor resistance genes, particularly inducible AmpC β-lactamases. As opportunistic pathogens, ECC strains are common clinical isolates, especially in patients admitted to intensive care units (ICUs). These pathogens often exhibit an MDR phenotype, which is mainly related to their ability to adapt to the hospital environment and acquire resistance genes encoded on mobile genetic elements (MGEs). The widespread and abused use of antimicrobial agents has also contributed to the emergence and spread of resistant pathogens [9]. This has made the treatment of bacterial infections increasingly difficult [10].

The crisis of bacterial resistance to β-lactam antibiotics traces its origins to ancient natural environments. Metagenomic studies reveal that genes encoding β-lactamases persist as an ancient reservoir of resistance genes within environmental microbes, and their original functions involve inter-microbial competition and signaling [11]. The extensive use of antibiotics in clinical and agricultural settings has exerted powerful selective pressure, mobilizing these naturally occurring resistance genes. Mobile genetic elements (MGEs), such as plasmids, transposons and integrons, play a pivotal role in this process. They capture intrinsic resistance genes from chromosomes (e.g., the bla_AMPC gene precursor in Enterobacter cloacae) and equip them with strong promoters, transforming them into highly expressible resistance determinants capable of horizontal transmission among diverse bacterial hosts [12]. Such plasmids carrying β-lactamase genes (e.g., bla_CTX−M, bla_NDM, bla_KPC, and so on) circulate efficiently among human, animal, and environmental microbial communities, leading to cross-border dissemination of resistance [13].

β-Lactamases degrade the β-lactam ring of β-lactam antimicrobial agents. They can be encoded by chromosomal or extrachromosomal elements. The class A serine β-lactamases are important enzyme sources of natural and acquired resistance to β-lactams, especially in gram-negative bacilli. To date, > 2,000 CABLs have been found, there are narrow-, broad-spectrum β-lactamases and carbapenemases, depending on the genetic determinants and they are highly diverse with many molecular variants. Different genotypes or variants may have different resistance spectra [14, 15]. Typical narrow-spectrum class A β-lactamases include TEM-1 and SHV-1. TEM-1 was the first TEM-type β-lactamase identified, with similar substrate and inhibition profiles as of SHV-1. TEM-1 production can confer up to 90% ampicillin resistance to E. coli and can hydrolyze ampicillin more rapidly than carbenicillin, cephalothin, or oxacillin. It shows mild activity against broad-spectrum cephalosporins and is inhibited by clavulanic acid. TEM-2 has an amino acid substitution with TEM-1 at position 39 (Gln39Lys), and the TEM-2 enzyme shows higher enzyme activity compared to strains producing TEM-1. Broad-spectrum β-lactamases may arise from point mutations of narrow-spectrum enzymes or, more commonly, from CTX-M-type ESBLs and they expand the spectrum of antibiotics that they can hydrolyze. Some variants of SHV family like SHV-3, SHV-5 and SHV-12 show increased resistance to ceftazidime. Higher resistance to cefotaxime than to ceftazidime is typical of all CTX-M-producing isolates. Finally, class A carbapenemases are exemplified by KPC variants, SME variants and some GES variants, and even rare CTX-M derivatives. They are capable of hydrolyzing carbapenem antibiotics like meropenem [16, 17].

Given the pivotal role of β-lactam antibiotics in anti-infective therapy, deciphering the evolutionary and transmission pathways of their resistance genes from environmental reservoirs to the pathogens holds critical scientific and clinical significance for early warning of emerging resistance threats and developing effective containment strategies. In this work, through molecular cloning and whole-genome sequencing, a novel class A β-lactamase gene, bla_KOB−1, was identified from the chromosome of an E. kobei isolate, and the structure of the novel resistance gene-related sequence was also characterized.

Materials and methods

Collection and identification of bacterial isolates

To learn the antimicrobial resistance status of the bacteria of animals and environment, a bacterial resistance survey project has been carried out in Wenzhou, China, and more than 500 bacteria were isolated from animals (including E. kobei DW108, which was isolated in this work from goose anal feces), and wastewater discharged from a livestock farm. Specimens from both animal feces and wastewater were collected by sterile cotton swabs. When sampling the specimens, the swab was inserted 3–5 cm into the anus of an animal and gently rotated or dipped into the wastewater and then the swab was immediately placed into a sterilized screw-cap specimen collection tube containing about 1 mL saline (0.9% NaCl). Single colonies were isolated by the streak plate method using Luria–Bertani agar plates, and colonies with different phenotypes were randomly collected. The genome of each isolate was sequenced on an Illumina HiSeq 2500 (Shanghai Personal Biotechnology Co., Ltd., Shanghai, China). Species identification was initially performed on the basis of 16S rRNA gene homology and then by genome-wide average nucleotide identity (ANI) as well as by digital DNA‒DNA hybridization (dDDH) analyses. 16 S rRNA gene homology was analyzed according to previously reported methods [18]. The genome-wide ANI was calculated by FastANI v1.33 [19], and dDDH was analyzed using the TYGS website (https://tygs.dsmz.de/). Table 1 lists the plasmids and bacteria used in this study.

Table 1.

Bacterial strains and plasmids used in this work

Strains and plasmids	Description	Source
DW108	The wild-type strain of E. kobei DW108	This work
DH5α	E. coli DH5α as a host for cloning of the blaKOB−1 gene	Our laboratory collection
BL21	E. coli BL21 as a host for expression of the blaKOB−1 gene	Our laboratory collection
ATCC 25922	E. coli ATCC 25922 as quality control strain for antimicrobial susceptibility testing	Our laboratory collection
pUCP24-blaKOB−1/DH5α	DH5α carrying the recombinant plasmid pUCP24-blaKOB−1	This work constructed
pCold I- blaKOB−1/BL21	BL21 carrying the recombinant plasmid pCold I-blaKOB−1	This work constructed
pUCP24	Cloning vector for the PCR product of the blaKOB−1 gene with its upstream promoter region, GENr	Our laboratory collection
pCold I	Expression vector for the PCR products of the ORF of the blaKOB−1 gene, AMPr	Our laboratory collection

^rResistance, GEN Gentamicin, AMP Ampicillin, ORF Open reading frame

Antimicrobial susceptibility testing

The minimum inhibitory concentrations (MICs) were determined using the standard agar dilution method on Mueller–Hinton (M-H) agar based on the Clinical and Laboratory Standards Institute (CLSI) guidelines, and all the values are averages of three independent tests. The breakpoints defined by CLSI M100 (34th Edition, 2024) were applied. E. coli ATCC 25922 was used as a quality control reference. The antimicrobial agents tested in this work (Table 2) were all purchased from Sinopharm chemical Peagent Co., Ltd. (Shanghai, China).

Table 2.

MICs of antimicrobial agents of the wild strains, the Recombinant strains carrying bla_KOB−1, bla_RUB−1 or bla_GIL−1, and the control strains (µg/mL)

Class	Antibiotic	Bla KOB−1					Bla RUB−1				Bla GIL−1			Susceptible (S)	Intermediate (I)	Resistant (R)
		E. kobei DW108	pUCP24-blaKOB−1/DH5α	pUCP24/DH5α	E. coli DH5α	ATCC25922	S. rubidaea 104261	S. rubidaea 103234T	E. coli DH10B(pRUB-1)	E. coli DH10B	C. gillenii CIP 106783T	E. coli DH10B (pGIL-1)	E. coli DH10B
β-Lactams	Ampicillin	64	16	4	4	4	-a	-	-	-	-	-	-	≤ 8	16	≥ 32
	Amoxicillin	128	64	2	4	8	128	128	> 512	2	16	> 512	2	≤ 8	16	≥ 32
	Ticarcillin	-	-	-	-	-	512	512	> 512	1	64	> 512	1
	Piperacillin	8	64	8	8	8	64	64	> 512	1	2	32	1	≤ 16	32–64	≥ 128
	Cephalothin	-	-	-	-	-	> 512	> 512	> 512	2	2	8	2
	Cefuroxime	-	-	-	-	-	64	64	16	2	2	8	2
	Cefazolin	4	0.5	0.5	1	2	-	-	-	-	-	-	-	≤ 2	4	≥ 8
	Cefoxitin	4	2	2	2	2	64	64	2	2	1	2	2	≤ 8	16	≥ 32
	Ceftazidime	0.25	0.06	0.06	0.06	0.125	1	1	2	0.06	0.06	0.5	0.06	≤ 4	8	≥ 16
	Cefotaxime	0.125	0.06	0.06	0.06	0.06	0.5	0.5	1	0.06	0.06	1	0.06	≤ 1	2	≥ 4
	Ceftriaxone	0.06	0.06	0.06	0.06	0.06	-	-	-	-	0.06	0.06	0.06	≤ 1	2	≥ 4
	Cefepime	0.015	0.015	0.015	0.03	0.03	0.5	0.5	0.5	0.06	0.06	0.12	0.06	≤ 2	-	≥ 16
	Cefpirome	-	-	-	-	-	0.25	0.25	0.5	0.06	0.06	0.25	0.06
	Moxalactam	-	-	-	-	-	0.5	0.5	0.5	0.12	0.12	0.25	0.12	≤ 8	16–32	≥ 64
	Aztreonam	< 0.03	0.06	0.06	0.06	0.06	0.25	0.25	1	0.06	0.06	0.12	0.06	≤ 4	8	≥ 16
	Imipenem	0.5	0.125	0.125	0.25	0.125	0.12	0.12	0.12	0.06	0.06	0.12	0.06	≤ 1	2	≥ 4
	Meropenem	0.125	< 0.004	0.008	0.008	0.008	-	-	-	-	-	-	-	≤ 1	2	≥ 4
β-Lactamase inhibitors	Piperacillin/Tazobactam	< 1/4	2/4	< 1/4	2/4	4/4	4/4	4/4	32/4	1/4	1/4	1/4	1/4	≤ 16/4	32/4–64/4	≥ 128/4
	Ceftazidime/Avibactam	0.0625/4	0.125/4	0.25/4	0.125/4	0.125/4								≤ 8/4	-	≥ 16/4
	Aztreonam/Avibactam	< 0.03/4	0.03/4	< 0.015/4	0.03/4	0.06/4	-	-	-	-	-	-	-	≤ 4/4		≥ 8/4
	Amoxicillin/Clavulanic acid	2/1	4/2	8/4	4/2	16/8	8/4	8/4	16/8	2/1	0.5/0.25	4/2	2/1	≤ 8/4	16/8	≥ 32/16
	Ticarcillin/ Clavulanic acid	-	-	-	-	-	8/2	8/2	32/2	18/2	1/2	8/2	1/2	≤ 16/2	32/2–64/2	≥ 128/2
Aminoglycosides	Gentamicin	0.5	-	-	-	0.5	-	-	-	-	-	-	-
	Tobramycin	0.25	-	-	-	0.5	-	-	-	-	-	-	-
	Kanamycin	1	-	-	-	2	-	-	-	-	-	-	-
	Ribostamycin	4	-	-	-	4	-	-	-	-	-	-	-
	Nalidixic acid	16	-	-	-	< 4	-	-	-	-	-	-	-
Quinolones	Levofloxacin	< 0.5	-	-	-	< 0.5	-	-	-	-	-	-	-
	Chloramphenicol	16	-	-	-	2	-	-	-	-	-	-	-

^aMIC not determined

Cloning of the bla_KOB−1gene and expression of KOB-1

To characterize the function of the predicted resistance gene, the gene with its promoter region was amplified via PCR using 2× Phanta^® Max Master Mix (Nanjing Vazyme Biotech Co., Ltd., Nanjing, China) with the in-house-designed primers pr-bla_KOB−1-XbaI-F and pr-bla_{KOB− 1}-BamHI-R (Table S1), and the bacterial DNA template was extracted using the Generay Genomic DNA Miniprep Kit (Shanghai Generay Biotech Co., Ltd., Shanghai, China). The PCR product was subsequently inserted into the cloning vector pUCP24 with a T4 DNA ligase cloning kit (Takara Bio, Inc., Dalian, China). The recombinant plasmid (pUCP24-bla_{KOB− 1}) was transformed into E. coli DH5α by the calcium chloride method [20]. The transformant (pUCP24-bla_{KOB− 1}/E. coli DH5α) was selected on Luria–Bertani (LB) agar plates supplemented with gentamicin (40 µg/mL). The cloned fragment was ultimately verified by Sanger sequencing.

The open reading frame (ORF) of bla_KOB−1 was amplified by primers with BamHI/HindⅢ restriction sites and a thrombin cleavage site (Table S1) and subsequently inserted into the expression vector pCold I. The recombinant plasmid (pCold I-bla_KOB−1) was subsequently introduced into E. coli BL21. The transformants were screened on LB agar plates supplemented with ampicillin (100 µg/mL), and the resulting fragments were further verified by Sanger sequencing. The transformant was cultured overnight in LB liquid medium, added to the medium in a conical flask at 1:100 and incubated until the OD₆₀₀ reached 0.6–0.8. The culture was incubated in an ice bath for half an hour, after which 1 mM isopropyl-β-dithiogalactopyranoside (IPTG; Sigma Chemicals Co., St. Louis, United States) was added to induce the expression of KOB-1. The cells were continuously cultured at 16 °C for 20 h. After induction by IPTG and then sonication, β-lactamase KOB-1 was purified with preequilibrated nickel-nitrilotriacetic acid (Ni-NTA) agarose resin (Beyotime Biotechnology, Shanghai, China), and the His-tag was subsequently removed by incubation with thrombin (Solarbio Science & Technology Co., Ltd., Beijing, China) for 48 h. The purity of KOB-1 protein was analyzed by SDS‒PAGE using 12% acrylamide separating gels and Coomassie blue G-250 staining. The protein concentration was assayed with a BCA Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL, United States).

Enzyme kinetic parameter analysis

The kinetic parameters of β-lactamase KOB-1 for the β-lactam substrates were tested on a UV–VIS spectrophotometer (SpectraMax M5; Molecular Devices, USA) in 10 mM phosphate buffer (pH 7.4) in a final reaction volume of 200 µL, and the hydrolysis measurement wavelengths and molar extinction coefficients for different substrates are listed in Supplementary Material TableS2. The steady-state kinetic parameters (k_cat and K_m) were calculated by nonlinear regression of the initial reaction rates using the Michaelis‒Menten equation in Prism (version 8.0.2) software (GraphPad Software, CA, United States).

Inhibition studies

The enzyme was preincubated with β-lactamase inhibitors (avibactam or tazobactam) for 5 min at 37 °C, and the concentrations of the inhibitors resulting in a 50% reduction in nitrocefin hydrolysis (IC₅₀) were determined using 100 nM nitrocefin as a substrate. The IC₅₀ values were obtained by nonlinear regression analysis (GraphPad Prism, version 8.0.2) using a log (inhibitor) vs. response (three parameters) curve [21]. The result is the average of three independent measurements.

Whole-genome sequencing (WGS) and sequence analysis

The bacterial genomic DNA of DW108 was extracted using the method mentioned above and sequenced by the Illumina HiSeq-2500 and PacBio RS II platforms at Shanghai Personal Biotechnology Co., Ltd. (Shanghai, China), with read lengths of PE150 and 10 kb, respectively, and a sequencing depth of 150×. The long reads from PacBio sequencing were assembled by Unicycler v0.4.8 [22]. The quality of the draft genome sequence was improved by mapping Illumina short reads onto the assembly using Pilon v1.23 [23]. Potential ORFs of the genome sequence were predicted by using Prokka v1.14.6 [24], and the functional annotation of these proteins was determined by DIAMOND v2.0.11 [25] and BLAST analysis against the Nonredundant Protein Sequence (NR) database of NCBI (https://www.ncbi.nlm.nih.gov/) and the UniProt/Swiss-Prot database at an e-value threshold of 1e-5. The resistance gene identifier v5.2.02 and the Comprehensive Antibiotic Resistance Database (CARD; McArthur et al., 2013) were used to predict the resistance genes. ISFinder (https://isfinder.biotoul.fr/) was applied to identify mobile genetic elements (MGEs) [26]. The phylogenetic tree of KOB-1 with the function-characterized class A β-lactamases of high identity was reconstructed by means of MEGA11 [27] and visualized using iTol [28]. Multiple sequence alignment of the protein sequences was conducted using MAFFT v7.490 [29], and the results were subsequently checked and filtered using the msa R package [30]. A map illustrating the genetic context of the resistance genes was generated with clinker v0.0.24 [31]. The molecular weight and pI of each protein were predicted using ProtParam (https://web.ExPASy.org/protparam/) [32].

Results and discussion

Resistance profile of the isolate DW108 and functional determination of the novel resistance gene bla_KOB−1

When analyzing the resistance phenotypes and genotypes of the bacteria, we found that one isolate, DW108, suggested the presence of a narrow-spectrum penicillinase (resistant to ampicillin and amoxicillin). Similar reports also revealed that three clinical E. kobei isolates exhibited increased MICs of ampicillin and piperacillin [2]. DW108 had high MICs of amoxicillin (128 µg/mL), ampicillin (64 µg/mL), but no functionally characterized β-lactam resistance gene was annotated from its genome. To determine the β-lactam resistance mechanism of DW108, we checked the annotation results of the genome and identified a chromosome-encoded hypothetical β-lactamase gene (designated bla_KOB−1 in this work) (Table S3). KOB-1 shared the highest amino acid similarity of 73.94% (100% coverage and 73.94% identity) with the functionally characterized class A β-lactamase RUB-1. To determine the function of the gene, bla_KOB−1 was cloned, and its resistance characteristics were analyzed.

During MIC testing, the clone harboring the bla_KOB−1 gene exhibited resistance to several antimicrobial agents. The MICs of amoxicillin, piperacillin and ampicillin were 32-, 8- and 4-fold greater, respectively, in the pUCP24-bla_KOB−1/DH5α recombinant strain than in the control strain, E. coli DH5α, without a significant difference in susceptibility to cephalosporin antibiotics. In addition, amoxicillin and piperacillin resistance were inhibited by the β-lactamase inhibitors clavulanic acid and tazobactam, respectively. However, the wild-type E. kobei DW108 exhibits resistance to ampicillin and amoxicillin, but is sensitive to piperacillin (Table 2). Possible reasons include low enzyme activity or the bacteria’s inherent sensitivity to the drug. It also may be attributed to the recombinant plasmid vector possessing a strong promoter, whereas the resistance gene in the wild-type strain is regulated by a weak promoter [33].

Among the functionally characterized β-lactamases, KOB-1 had the highest amino acid sequence identity to RUB-1 (73.94%), followed by GIL-1 (72.89%), but their resistance profiles differed. The β-lactam resistance spectrum of bla_KOB−1 decreased. According to the previous publications, it seemed that the β-lactam resistance spectra of the cloned genes bla_RUB−1 and bla_GIL−1 were wider than that of bla_KOB−1. Compared with their controls, the recombinant strain carrying the bla_KOB−1 gene only showed resistance to piperacillin and amoxicillin, whereas bla_RUB−1 and bla_GIL−1 not only showed resistance to the two antibiotics but also showed resistance to the other two antimicrobials cefotaxime and ceftazidime [34, 35].

Taxonomic classification and genomic characteristics of isolate DW108

The 16S rRNA gene sequence of strain DW108 showed 100.0% coverage and 100.0% identity to the sequence of E. kobei JCM 8580^T (NR_113321.1), the type strain of the species E. kobei (Table S4). Among all the genomes of E. kobei in the NCBI database, DW108 also had the highest ANI (98.44%) with E. kobei JCM 8580^T (GCA_018323985.1), which is above the ≥ 95% ANI cutoff for defining a bacterial species (Table S5). The dDDH results calculated on the TYGS website showed that DW108 had a greater dDDH value (88.70%) with E. kobei ATCC BAA-260^T (the JCM Strain Preservation Centre number for E. kobei ATCC BAA-260^T is E. kobei JCM 8580; thus, E. kobei ATCC BAA-260^T and E. kobei JCM 8580 are the same), which was above the 70% cutoff for defining a bacterial species (Table S6) [36]. As a result, DW108 was ultimately classified as the species E. kobei and named E. kobei DW108.

The genome of E. kobei DW108 comprises a chromosome and three plasmids. The chromosome was approximately 4.60 Mb in size and encoded 4,385 ORFs with a GC content of 55.36%, while the three plasmids designated pDW108-119542, pDW108-58662 and pDW108-4570 were 119.5, 58.7 and 4.6 kb in size, encoding 140, 80 and 8 proteins, respectively (Table 3). A total of 260 assembled and annotated genomes of E. kobei were available in the NCBI database, with a median length of 4.95 Mb, median ORFs of 4,566, and a median GC content of 54.81%; 18 of these genomes carried 1 to 12 plasmids, with the smallest being 2.02 kb (plasmid pIB2020_S from E. kobei IB2020, NZ_CP059486.1) and the largest being 282.90 kb (plasmid pSTN0717-53-1 from E. kobei STN0717-53, NZ_AP022499.1) in length. The size and encoded ORFs as well as the GC content of the E. kobei DW108 genome were smaller than the species average. The sizes of the three plasmids of E. kobei DW108 are intermediate to those of the E. kobei plasmids in the NCBI database.

Table 3.

General features of the E. kobei DW108 genome

	Chromosome	pDW108-119542	pDW108-58662	pDW108-4570
Size (bp)	4,595,692	119,542	58,662	4,570
GC content (%)	55.36	50.99	54.24	47.05
Predicted coding sequences (CDSs) of proteins	4,385	140	80	8
Known proteins	3,805 (86.77%)	73 (52.14%)	56 (70%)	1 (12.5%)
Hypothetical proteins	580 (13.23%)	67 (47.86%)	24 (30%)	7 (87.5%)
Protein coding (%)	96.52	98.59	94.12	100
Average ORF length (bp)	928.1	732.5	620.9	416.3
Average protein length (aa)	310.2	244.8	207.8	137.9
tRNAs	89	0	0	0
rRNA operons	(16S–23S-5S) × 5, (16S–23S-5S–5S) × 1, (16S–23S) × 1	0	0	0

Enzyme kinetic parameters of KOB-1

To analyze the kinetic parameters of the novel β-lactamase, the protein KOB-1 with a molecular weight of 30.52 kDa was expressed and purified (Figure S1). It could hydrolyze penicillin antibiotics (penicillin G, ampicillin and amoxicillin) and some cephalosporins, such as cefazolin and cefotaxime; however, its hydrolysis of cefoxitin or aztreonam was not detected. Of the substrates analyzed, KOB-1 exhibited the strongest hydrolytic activity against penicillin G (k_cat/K_m = 275.05 ± 15.08 mM^{− 1}·s^{− 1}) but weaker hydrolytic activity against cefotaxime (k_cat/K_m = 0.31 ± 0.03 mM^{− 1}·s^{− 1}) (Table 4). KOB-1 hydrolyzed cefotaxime very inefficiently, especially compared to the other variants. The MIC for the recombinant strain harboring bla_KOB−1 did not significantly differ from that of the control, i.e., it did not show significant resistance to cefotaxime in vivo. This phenomenon has been reported previously, where BAT-2 and BSU-2 exhibited slight hydrolytic activity against ampicillin, but in vivo antimicrobial resistance was not detected for either strain [37]. The underlying reason for this phenomenon may be (1) Metabolic Burden: The high cost of maintaining the plasmid and expressing the foreign protein drains cellular energy and resources. This weakens the cell overall, making it more susceptible to antibiotic stress, even if the resistance enzyme is active. The protection offered by the enzyme is outweighed by the general vulnerability of the cell [38, 39]. (2) Trojan Horse Effect: The resistance enzyme captures the antibiotic but hydrolyzes it inefficiently or locally in the periplasm. This process creates a high local concentration of antibiotic or its intermediates near the cell wall synthesis targets, inadvertently intensifying the damage and increasing sensitivity [40]. (3) Protein Mislocalization or Misfolding: Due to signal peptide issues, the resistance enzyme may fail to be transported to the periplasmic space (mislocalization), preventing it from providing protection [41]. (4) Unintended Disruption of Cell Wall Integrity: High-level expression of the resistance protein itself may physically interfere with normal cell wall assembly, directly compromising its structural integrity and rendering the cell more susceptible to the antibiotic [42]. According to the reports, KOB-1, RUB-1 and GIL-1 all exhibited hydrolytic activity against amoxicillin and cefotaxime, but neither of them showed hydrolytic activity against cefoxitin. This finding is consistent with the difference in MICs of the antimicrobial agents. The ability of GIL-1 to hydrolyze other substrates (including ampicillin, penicillin G and cefazolin) that KOB-1 does has not been reported [34]. Both avibactam and tazobactam had strong inhibitory effects (IC₅₀ values of 7.31 nM and 34.29 nM, respectively) and clavulanic acid showed a weaker inhibitory effect of 148.07 nM, relatively on KOB-1 when the concentrations of KOB-1 and nitrocefin were 28.39 nM and 100 nM, respectively. This result is consistent with the reported inhibitory activities of β-lactamase inhibitors against class A β-lactamases [35].

Table 4.

Kinetic parameters of KOB-1, RUB-1 and GIL-1 for β-lactam antibiotics

Enzyme	KOB-1			RUB-1d			GIL-1e
Substrate	kcat (s− 1) a	Km (µM) a	kcat/Km(mM− 1·s− 1)a	kcat (s− 1) a	Km (µM) a	kcat/Km(mM− 1·s− 1)a	kcat (s− 1) a	Km (µM) a	kcat/Km(mM− 1·s− 1)a
Ampicillin	5.84 ± 1.21	945.30 ± 26.17	6.21 ± 1.42	-	-	-	-	-	-
Amoxicillin	10.42 ± 0.67	437.13 ± 18.47	23.81 ± 0.52	80.9	55	1,471	381	84	4,540
Penicillin G	43.34 ± 1.43	157.87 ± 6.51	275.05 ± 15.08	-	-	-	-	-	-
Cefazolin	1.36 ± 0.24	145.67 ± 9.64	9.27 ± 1.13	-	-	-	-	-	-
Cefotaxime	0.11 ± 0.01	337.73 ± 11.39	0.31 ± 0.03	7.7	155	49.6	7	714	10
Cefoxitin	NAb	NAb	NAb	NAb	NAb	NAb	NAb	NAb	NAb
Aztreonam	NAb	NAb	NAb	NDc	NDc	NDc	0.8	938	1
Benzylpenicillin	-	-	-	252.4	337	749	286	51	5,600
Ticarcillin	-	-	-	17.3	129	134	100	10	10,000
Piperacillin	-	-	-	32.9	27	1,219	306	90	3,400
Cephalothin	-	-	-	19	200	94.6	37	62	598
Cefuroxime	-	-	-	8.6	813	10.6	2	88	23
Ceftazidime	-	-	-	NDc	NDc	NDc	1	1,749	0.6
Cefepime	-	-	-	2.2	176	12.5	77	2,685	29
Imipenem	-	-	-	6.2	392	15.7	0.2	2,977	0.05

^aThe values are the mean ± standard deviation

^bNA, no hydrolytic activity detected

^cND, not determinable because the Km values were too high

^dReferred from [35]

^eReferred from [34]

Evolutionary relationships of KOB-1 with other functionally characterized β-lactamases

An evolutionary tree of the functionally characterized β-lactamases with greater than 70.0% identity with KOB-1 from the Beta-Lactamase Database (http://www.bldb.eu/) and the CARD database (https://card.mcmaster.ca/) was constructed (Fig. 1). All the genes encoded class A β-lactamases. The novel β-lactamase KOB-1 was located closer to the β-lactamases that shared high identity with KOB-1, such as RUB-1 (73.94%, the highest) and GIL-1 (72.89%). These β-lactamases were all found in bacteria of different genera; KOB-1 was initially identified in E. kobei; RUB-1 and GIL-1 were originally isolated from Serratia rubidaea and Citrobacter gillenii, respectively; and the rest were TEM-type enzymes initially isolated from E. coli with the exceptions of TEM-169 and TEM-1 which were first isolated from K. pneumoniae and A. baumannii. KOB-1 formed a new branch in the evolutionary tree, and all the other proteins were more distantly related to KOB-1.

Fig. 1.

A phylogenetic tree showing the evolutionary relationship of KOB-1 with other functionally characterized class A β-lactamases. KOB-1 from this study is highlighted in red

Conservation of the functionally essential residues of the β-lactamase KOB-1

Multiple sequence alignments of KOB-1 with other class A β-lactamases reported previously revealed that they all contained a serine active site, SXXK (serine-X-X-lysine), at position 66–69 and three additional conserved motifs for class A β-lactamases, namely, SDN (serine-aspartic acid-asparagine) at position 126–128, KTG (lysine-threonine-glycine) at position 230–232 and EXXLN (glutamate-X-X-leucine-asparagine) at position 162–166, where Ser66 is the key catalytic residue, while Lys69 and Glu162 serve as general bases for the acylation and deacylation steps, respectively [43]. One of the conserved residues of KOB-1 and GIL-1 is threonine at position 231, whereas the other TEM enzymes listed in Fig. 2 are serine. In addition, three major motifs (S66TFKVL, S126DNTAANL, and R160XEXXLN) involved in the catalytic mechanism and/or substrate binding of the enzymes of the A1 subclass of class A β-lactamases were found [15] (Fig. 2).

Fig. 2.

Multiple alignment of the amino acid sequences of KOB-1 and other functionally characterized β-lactamases. Exclamation marks indicate fully conserved residues, asterisks indicate strongly similar residues, and gaps are represented using hyphens. The numbers on the right represent the corresponding sequence length. The characteristic serine active site of the β-lactamase is boxed in red. Three conserved motifs of class A β-lactamases are boxed in orange. The three major motifs of the A1 subclass enzymes of class A β-lactamases are marked at the appropriate location. KOB-1 is the protein from this study and is shown in red

Genetic context of the bla_KOB−1 gene

The structure of the bla_KOB−1 gene-related sequences was analyzed. Using the bla_KOB−1 gene as a query to search for similar genes with > 80.0% global nucleotide identity in the nucleotide database of NCBI by BLASTN, only three bla-like genes that shared nucleotide similarities of 97.89% (AP024590.1), 85.21% (CP019113.1) and 85.21% (CP127837.1) with bla_KOB−1 were found. Of the three bla-like genes, one (E. kobei JCM 8580, AP024590.1) was from the same species as bla_KOB−1, and the other two with unclassified species (Enterobacter sp. SA187, CP019113.1 and Enterobacter sp. PGRG2, CP127837) were from the same genus Enterobacter as bla_KOB−1 (Fig. 3). The corresponding proteins they encoded, such as BH713_13970 (OLR17638) encoded by E. kobei JCM 8580 and the unnamed class A β-lactamase (WP_072569066) encoded by Enterobacter sp. SA187 with 98.0% and 89.0% aa identity with KOB-1, respectively, might have the functions similar to those of KOB-1. When the 20 kb sequence with the bla_KOB−1 gene at the center was used as a query instead, nine sequences with > 80.0% global nucleotide identity were retrieved from the nucleotide database of NCBI, including three containing the bla-like genes mentioned above. A comparative analysis of the 10 sequences (including one of this work) revealed that, except for a fragment encoding two genes (bla-phoX), which exists in only four sequences, the 10 sequences share nearly the same genetic context in terms of gene content or gene order. Bla-phoX has been found only in Enterobacter species, and was not related with any mobile genetic element (MGE), which suggests that it is intrinsically conserved in Enterobacter. In contrast to the four sequences with a bla gene, the six sequences lacking a bla gene were all from the genus Leclercia. In addition, no mobile genetic element was identified in the flanking regions of the bla_KOB−1(-like) genes, although an IS481 was found to be inserted between terC and asmA in one sequence (NZ_CP035382) which was free of a bla gene. These findings suggested that the bla_KOB−1(-like) genes were relatively conserved in the bacteria of the genus Enterobacter of the family Enterobacteriaceae (Fig. 3). Enterobacteriaceae bacteria serve as a reservoir for certain endogenous β-lactamases, such as the naturally occurring β-lactamase MIR-1 in Enterobacter cloacae [44].

Fig. 3.

Genetic environment of bla_KOB-1 and its close relatives. The orientation of the genes is shown via an arrow. The bla_KOB-1 gene and putative class A β-lactamase (bla-like) genes are colored red. The genes upstream of the bla_KOB-1 (bla-like) gene are colored blue, and the downstream genes are colored yellow. orfA is a polysaccharide export protein-encoding gene. Regions with 60%-80% amino acid identity are colored light gray, regions with 80%-100% amino acid identity are colored dark gray

Conclusion

In this study, we characterized a novel class A β-lactamase gene, bla_KOB−1, which is encoded on the chromosome of E. kobei DW108 and confers resistance to penicillin antibiotics. Like other class A β-lactamases, clavulanic acid, avibactam and tazobactam are strong inhibitors of this novel β-lactamase. bla_KOB−1-like genes have been identified in different strains of the genus Enterobacter. Not carried by a MGE, the bla_KOB−1 gene represents an intrinsic resistance mechanism. Discovery of new resistance mechanism may help clinic to treat the infectious diseases caused by the bacteria carrying this resistance gene more effectively.

Abbreviations

MIC

Minimum inhibitory concentration

CARD

Comprehensive antibiotic resistance database

ECC

Enterobacter cloacae complex

ICU

Intensive care units

MDR

Multi-drug resistance

MGEs

Mobile genetic elements

CABL

Class A serine β-lactamase

ANI

Average nucleotide identity

dDDH

Digital DNA–DNA hybridization

GEN

Gentamicin

AMP

Ampicillin

MH

Mueller-Hinton

CLSI

Clinical and Laboratory Standards Institute

PCR

Polymerase chain reaction

ORF

Open reading frame

OD

Optical density

IPTG

Isopropyl-β-d-thiogalactopyranoside

SDS‒PAGE

Sodium dodecyl sulfate‒polyacrylamide gel electrophoresis

IC50

Half maximal inhibitory concentration

WGS

Whole-genome sequencing

NCBI

National center for biotechnology information

CDSs

Coding sequences

Authors’ contributions

Conceived and designed the experiments: QB, TX, HZ and XZ; Performed the experiments: JZ, JL, CS, YZ and CL; Data analysis and interpretation: JZ, QY and JL; Drafting of the manuscript: JZ, QB, TX and XZ.

Funding

This study was supported by the Science & Technology Project of Jinhua City, China (2024-4-030, 2022-2-013), Major scientific and technological innovation project of Wenzhou, China (ZY2022013) and the Science & Technology Project of Wenzhou City, China (N20210001).

Data availability

The chromosome, three plasmids and bla_KOB-1 gene sequences of E. kobei DW108 have been deposited in GenBank. The nucleotide sequence accession numbers are CP135147 (chromosome), CP135148 (pDW108-119542), CP135149 (pDW108-58662), CP135150 (pDW108-4570) and OR855418.1 ( bla_KOB-1 ). The data is publicly available.

Declarations

Ethics approval and consent to participate

This study was conducted according to the principles of the Declaration of Helsinki. This study used strains isolated from the environment and animals in animal farms in Wenzhou, China. The specimens we collected were obtained with the owner’s consent. The owners of the farms were informed in writing of the study and provided approval for the sampling of animals. The studies involving human participants and animals were reviewed and approved by the Animal Welfare and Ethics Committee of Wenzhou Medical University, Zhejiang Province, China (protocol number: wydw2021-0323). All authors have reviewed and approved the manuscript for publication. As our study is not a clinical trial, no clinical trial registration or approval is required. Clinical trial registration number: Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.