Tracing the possible evolutionary trends of Morganella morganii: insights from molecular epidemiology and phylogenetic analysis

ABSTRACT Morganella morganii, encompassing two subspecies, subsp. morganii and subsp. sibonii, is a common opportunistic pathogen, notable for intrinsic resistance to multiple antimicrobial agents. Despite its clinical significance, research into the potential evolutionary dynamics of M. morganii remains limited. This study involved the analysis of genome sequences from 431 M. morganii isolates, comprising 206 isolates that cause host infections, obtained from this study and 225 from the NCBI genome data sets. A diverse array of antimicrobial resistance genes (ARGs) was identified in M. morganii isolates, including mcr-1, tet(X4), tmexCD-toprJ, and various carbapenemase genes. In addition, a novel blaKPC-2-bearing plasmid with demonstrated conjugative capability was discovered in M. morganii. The majority of virulence-related genes (VRGs), except for the hlyCABD gene cluster, were found in almost all M. morganii. Three novel genospecies of M. morganii were identified, designated as M. chanii, M. variant1, and M. variant2. Compared to M. sibonii, M. chanii genospecies possessed a greater number of flagellar-related genes, typically located within mobile genetic elements (MGEs), suggesting potential for better environmental adaptability. Phylogenetic analysis further disclosed that M. morganii was divided into 12 sequence clusters (SCs). Particularly, SC9 harbored an elevated abundance of ARGs and VRGs, mainly toxin-related genes, and was associated with a higher presence of MGEs compared to non-SC9 strains. The collective findings suggest that M. morganii undergoes evolution driven by the influence of MGEs, thereby significantly enhancing its adaptability to selective pressures of environmental changes and clinical antimicrobial agents. IMPORTANCE The growing clinical significance of Morganella morganii arises from its abundant virulence factors and antimicrobial resistance genes, resulting in elevated infection rates and increased clinical scrutiny. However, research on the molecular epidemiology and evolutionary trends of M. morganii has been scarce. Our study established a list of virulence-related genes (VRGs) for M. morganii and conducted a large-scale epidemiological investigation into these VRGs. Based on genomic classification, three novel genotypes of M. morganii were identified, representing evolutionary adaptations and responses to environmental challenges. Furthermore, we discovered the emergence of a sequence cluster enriched with antimicrobial resistance genes, VRGs, and mobile genetic elements, attributed to the selective pressure of antimicrobial agents. In addition, we identified a novel conjugative plasmid harboring the blaKPC-2 gene. These findings hold significance in monitoring and comprehending the epidemiology of M. morganii.

M organella morganii, a facultative anaerobic Gram-negative bacterium, is ubiqui tously found in environmental habitats and the gastrointestinal tracts of humans, mammals, and reptiles (1).M. morganii is categorized into two subspecies, M. morganii subsp.sibonii and M. morganii subsp.morganii, according to trehalose fermentation ability which is linked to the presence of the trehalose operon (treR, treB, and treP) (2).M. morganii is recognized as a significant opportunistic pathogen, frequently identified as a causative agent of nosocomial infections in adults, notably contributing to cases of urinary tract and wound infections during clinical observations (3).The mortality of M. morganii infections remained elevated in reported cases.A prior study showed a total of 150 out of 709 patients (21.2%) succumbed within 30 days following M. morganii bloodstream infections (BSIs) (4).
M. morganii possesses intrinsic resistance to a broad spectrum of antimicrobial agents including ampicillin, amoxicillin, and most of the first-and second-generation cephalosporins, attributed to chromosomally encoded bla AmpC (5) as well as macrolides, lincosamides, glycopeptides, fosfomycin, fusidic acid, and colistin (6).Akin to other Enterobacterales, M. morganii also harbors a diverse array of acquired antimicrobial resistance genes (ARGs) such as bla KPC-2 , bla NDM , tet(A), and aadA1, facilitated by the substantial presence of mobile genetic elements (MGEs) (7).Moreover, various viru lence-related genes (VRGs) such as urease, iron acquisition systems, IgA protease, and hemolysins were identified in M. morganii, characterizing its ability to colonize diverse hosts.Notably, comparative genome analysis revealed that unique VRGs such as eut operon were carried by the M. morganii genome compared to other Proteeae members (8).Therefore, M. morganii has been labeled an emerging "superbug." However, up to this point, there has been a lack of large-scale and long-term investigation into the carriage of VRGs in M. morganii.
Molecular epidemiology plays an integral role in understanding pathogenic bacteria, offering insights into their transmission routes, virulence, drug resistance, genetic variations, and evolutionary trends.Among various molecular techniques, multilocus sequence typing (MLST) stands out as a versatile and universally applicable sequencebased methodology, instrumental in elucidating clonal relationships among bacterial species (9).Regrettably, the MLST typing method has not yet been established for M. morganii.Currently, only a limited number of studies have conducted genomic epidemiology analysis on M. morganii using whole-genome data.Jing et al. developed a classification system based on genome sequencing, which categorized M. morganii into four distinct genospecies: M. morganii, M. chanii, M. sibonii, and M. laugraudii (7).However, the genomic characteristics of these individual genospecies were not delved into in detail.This study conducted the molecular epidemiological analysis of M. morganii including 206 isolates from this study and 225 isolates from the NCBI genome data sets and aimed to meticulously examine the characteristics of different sequence clusters, thereby elucidating the evolution of the M. morganii genome.

Antibiotic resistome, virulence, and MGEs characteristics of M. morganii
The analysis of the antibiotic resistome of 431 M. morganii isolates (206 from this study and 225 from NCBI genome data sets) revealed a large range of resistance determinants, with 110 kinds of ARGs conferring resistance to 15 different categories of antimicrobial agents.The ARG count per isolate varied from 0 to 55, averaging 6.3 genes per isolate (Table S2).Notably, ARGs involved in resistance to multiple "last resort antimicrobial agents" including carbapenems (bla KPC , bla NDM , bla IMP , bla GES , and bla OXA) , polymixins (mcr), and tigecycline [tet(X4), and tmexCD-toprJ] were identified in M. morganii (Table S2).In parallel, a detailed screening of 133 virulence-related genes (VRG) profiles was conducted in the 431 M. morganii isolates.The vast majority of VRGs, such as genes related to motility-chemotaxis systems, urease hydrolysis and putrescine production, and fimbrial adhesins, were found to be nearly ubiquitous among the isolates (Tables S1 and  S3).However, the hlyCABD operon, encoding the RTX toxin linked to bacterial hemoly sis, was identified in only 29.0% (125/431) of the isolates (Table S3) (10).The analysis identified diverse MGEs in M. morganii.In addition to numerous insertion sequences and transposons, M. morganii also possesses a high load of genomic islands (averaging 4.9 per strain) and prophages (averaging 5.8 per strain) (Table S2).The M. morganii harboring genes, including ARGs, VRGs, and MGEs, were compared according to their sources (Tables 2 and 3; Table S4).The results suggest that there were no significant differences in the quantities of ARGs, VRGs, and MGEs carried by M. morganii collected from various geographic locations in this study.Compared to M. morganii from the NCBI genome data sets, those collected for this study exhibited a higher prevalence of VRGs but a lower prevalence of MGEs (P < 0.05).Considering that these differences might stem from the diverse host spectrum in the NCBI genome data sets, further analysis was conducted on the gene carriage among M. morganii from different host sources.The results revealed that M. morganii from Homo sapiens carried the most VRGs (P < 0.05).

Emergence of a novel bla KPC-2 -bearing plasmid in M. morganii
To further understand the dissemination of carbapenemase genes in M. morganii, a detailed analysis was conducted on 37 fragments containing carbapenemase genes (14 bla KPC-2 , 13 bla NDM-1 , and 10 other types).In the carbapenemase gene at known locations, aside from bla IMP-27 and two bla NDM-1 located in the chromosome, all remaining carbapenemase genes were found on plasmids (Table 1).This strongly suggests that horizontal transmission of plasmids was the primary driving factor behind the prevalence of carbapenemase genes in M. morganii.While almost all of these plasmids were commonly encountered in other Enterobacteriales (11), a novel bla KPC-2 -harboring plasmid belongs to unknown Inc-type was identified in three isolates: 2387, GCF_018456265.1, and GCF_018456285.1, named separately as p2387-KPC-2 (CP139442.1),p229813-KPC (MN310368.1)and p516602-KPC (MN310367.1,linear and therefore not included in subsequent studies) (Table 1).When searched in the NCBI genome data sets, these plasmids were found to share a high degree of sequence homology with two M. morganii-hosted plasmids, p11759-FII (MZ848139.1)and p46903_2 (CP070522.1),neither of which contained carbapenemase genes.Compared to p11759-FII and p46903_2, the plasmids p2387-KPC-2 and p229813-KPC possessed an additional segment of 12,440 bp.This segment corresponded to the transposition unit featuring one transposase and two direct repeat sequences of CTGAAT (Fig. 2A).Upon further comparison, this transposition unit exhibited high similarity to a bla KPC-2bearing fragment of Escherichia coli YL03 (CP093551) in the NCBI genome data sets (Fig. 2B).This finding suggested that the transposition unit carrying bla KPC-2 and bla TEM-1B integrated into the M. morganii-hosted plasmid, resulting in the formation of a novel plasmid.Supporting this finding, the strain harboring p46903_2 was also found to possess another bla KPC-2 -bearing plasmid, named p46903-KPC (Table 1).Furthermore, conjugation experiments confirmed the transferability of this novel bla KPC-2 -bearing plasmid, with the conjugation frequency of 7.9*10 −5 when transferred to the EC600 strain, suggesting its potential for horizontal transmission among Enterobacteriales.

Emergence of a genospecies with abundant flagellar-related genes in M. morganii subsp. sibonii based on genomic classification
To conduct the genomic classification of M. morganii, we calculated pairwise average nucleotide identity (ANI) values for 431 isolates.All M. morganii isolates were classified into five distinct genospecies, adhering to the 95% ANI threshold for genospecies delineation (Fig. 3A) (12).Based on the previous research ( 7), these genospecies were designated as M. morganii (381 isolates), M. sibonii (36 isolates), M. chanii (12 isolates), M. variant1 (one isolate, corresponding to M. laugraudii in the previous research), and M. variant2 (one isolate, not identified in the previous research).In addition, a principal component analysis (PCA) plot similarly indicated that the 431 M. morganii could be categorized into five distinct clusters (Fig. 3B).To comprehend the distinctions in the trehalose operon (treRBP) among different genospecies excluding M. morganii and one M. chanii genospecies without treRBP, a total of 49 treRBP sequences were extracted for multiple sequence alignment and phylogenetic tree construction.The findings unveiled that the treRBP within the identical genospecies were categorized into corresponding clusters, consistent with prior ANI and PCA analyses (Fig. S1).Despite the sequence variations observed in the treRBP genes across different genospecies, all retained the functional capability to ferment trehalose.Thus, M. sibonii, M. chanii, M. variant1, and M. variant2 genospecies belonged to M. morganii subsp.sibonii, and M. morganii genospe cies fall under M. morganii subsp.Morganii, according to previous classification (13).
Building upon the background information of 431 isolates, we conducted a detailed analysis of the epidemiological and molecular characteristics of different M. morganii genospecies.While the M. morganii genospecies predominantly originated from Homo sapiens (87.9%), a significantly larger proportion of M. sibonii and M. chanii strains were derived from environmental sources, accounting for 38.9% and 50%, respectively.There was no significant variation in the geographical and temporal distributions of the different genospecies (Fig. 4A).Leaving aside M. variant1 and M. variant2 genospecies, which had only one isolate each, M. morganii genospecies exhibited the highest load of ARGs (averaging 6.7 per strain) and VRGs (averaging 130.0 per strain), while M. chanii genospecies displayed the fewest of these genes (averaging 1.4 ARGs and 120.7 VRGs per strain) (P < 0.05).Interestingly, M. chanii genospecies harbored an equivalent amount of MGEs (averaging 18.6 per strain) to that of M. morganii and sibonii (averaging 17.0 and 18.6 per strain, respectively) (Fig. S2A).A comparative genomic analysis between M. chanii and M. sibonii genospecies revealed a notable abundance of cell motility genes, primarily flagellar-related and predominantly located within MGEs, in the M. chanii genospecies (Table S5).Contrastingly, M. variant1 and M. variant2 genospecies harbored minimal ARGs (0 and 1, respectively) and VRGs (112 and 111, respectively), significantly lower than the average counts (6.3 of ARGs and 129.0 of VRGs) (Table S2).In stark contrast, they possessed 25 and 27 MGEs, respectively, with particular enrichment in genomic islands (14 in M. variant2) and prophages (eight for both M. variant1 and M. variant2) (Table S2).Therefore, M. variant1 and M. variant2 genospecies demonstrated a greater number of genes related to replication, recombination, repair, and transcription compared to M. sibonii and M. chanii (Tables S6 and S7).

Emergence of a sequence cluster enriched with ARGs, VRGs, and MGEs in M. morganii subsp. morganii based on phylogenomic analysis
For further phylogenomic analysis of the M. morganii, a maximum likelihood phylog eny was constructed based on 70311 filtered cgSNPs identified from 431 M. morganii sequences.The rhierBAPS population structure analysis clustered the 431 genomes into 12 distinct sequence clusters (SCs), denoted as SC1 through SC12.As expected, M. sibonii and M. chanii genospecies were divided into two SCs, SC1 and SC2, respectively (Fig. 4A).The M. variant1 and M. variant2, each comprising only one isolate, have not been categorized into any of the identified SCs.The presence of M. morganii from varied sources, geographic locations, and time points within the same SC suggested a widespread propagation capability of the strain across different hosts and regions.The distribution of sequence clusters among M. morganii from various geographic locations in this study shows no significant differences (Table S4).However, the distribution of sequence clusters SC1, SC2, and SC8 varied significantly among M. morganii from different hosts (Table 2, P < 0.05).In addition, significant differences were observed in the distribution of SC1, SC7, and SC11 between isolates from the NCBI genome data sets and this study (Table 3, P < 0.05).A noteworthy revelation lay in the strong correlation observed between the sequence cluster of M. morganii and the specific β-lactamase DHA (or MOR) types it inherently harbored (Fig. 4B).Within multiple SCs, the majority of strains predominantly produced a specific DHA type, for instance, SC1 strains predomi nantly featured DHA-16, while SC9 strains were characterized by DHA-4 production.
A comprehensive analysis was undertaken to assess the abundance of ARGs, VRGs, and MGEs within each SC of M. morganii.Overall, different categories of ARGs and MGEs, excluding prophage, exhibit significant variations across distinct clusters with a coefficient of variation greater than 15%.By contrast, VRGs displayed a relatively homogeneous distribution across the clusters, apart from genes related to toxins and the type III secretion system (Fig. 5B).Notably, one specific sequence cluster, SC9, classified under M. morganii genospecies, stood out with the highest average counts of VRGs and MGEs, and ranked second in ARG abundance, only surpassed by SC4 (Fig. 5A).The isolate GCF_016618235.1 in SC4 harbored 55 ARGs due to gene copies, elevating the average ARG count in SC4 (14).The ARGs, VRGs, and MGEs in SC9 strains were significantly more abundant than those in non-SC9 strains (P < 0.01, Fig. S2B).Focusing on ARGs, SC9 strains carried more genes conferring resistance to aminoglycosides, trimethoprims, sulfonamides, and tetracyclines than non-SC9 strains (P < 0.05, Fig. 5B).Specifically, the prevalence of 12 genes such as bla CARB-2 , tet(B) and tet(A) was higher in SC9 strains, while dfrA17, catA2, and aadA5 exhibited higher prevalence in non-SC9 strains (P < 0.05, Fig. 5C).In the realm of VRGs, SC9 strains displayed a marked disparity in toxin-related genes compared to non-SC9 strains, especially hlyCABD encoding RTX toxin (15).These gene clusters were present in 97.3% of SC9 strains and significantly less prevalent in 22.6% of non-SC9 strains (Fig. 5C).In addition, the abundance of sodC (related to superoxidestress) and btuB (involved in iron acquisition) was significantly higher in SC9 strains, while sctC (associated with type III secretion system) was more prevalent in non-SC9 strains (Fig. 5C).Except for insertion sequences and transposons, the number of other MGEs in SC9 strains was significantly higher than in non-SC9 strains.

DISCUSSION
MGEs, such as genomic islands (GIs), integrative conjugative elements (ICEs), transposes (Tns), insertion sequences (ISs), and prophages, were instrumental in markedly shaping bacterial genome structures (16).These elements facilitated the exchange of genetic material, thus playing a pivotal role in the evolution of bacterial strains (17).Prior research has identified the high prevalence of various MGEs in M. morganii, underlining their significant role in enhancing genomic plasticity within this species (7,14,18,19).
In line with previous reports (2, 7), our study detected five distinct genospecies within the M. morganii at the genome scale, surpassing two clusters that correspond to the two conventional subspecies.In addition, sequence variations in treRBP genes were observed among these genospecies of M. morganii subsp.sibonii, suggesting evolution ary divergence within the subspecies with a substantial proportion of isolates originating from the environment.Despite abundant MGEs, M. chanii genospecies demonstrated a notably lower prevalence of ARGs and VRGs compared to M. sibonii genospecies.
Comparative genomic analysis reveals that M. chanii genospecies carries a greater number of flagellar-related genes, primarily located within MGEs (Table S5).Considering that flagellar-related genes contributed to movement toward favorable environments and that environmental complexity played a crucial role in strain evolution (20,21), the presence of a diverse array of MGEs in M. chanii genospecies might facilitate the adaptive evolution in response to environmental challenges rather than an augmented repertoire of resistance and virulence determinants.Unlike M. chanii, M. variant1 and M. variant2 genospecies were found to lack a substantial number of flagellar-related genes.Instead, these genospecies were characterized by a significant presence of genes related to replication, recombination, repair, and transcription, which was attributed to their abundant GIs and prophages (Tables S6 and S7).Further research is warranted to determine whether these genomic differences manifest as distinct phenotypic charac teristics.Unlike M. morganii subsp.sibonii, which was primarily isolated from environmental sources, M. morganii subsp.morganii was predominantly isolated from Homo sapiens, and its evolutionary trajectory was significantly influenced by the selective pressure of antimicrobial agents.Phylogenomic analysis, integrated with molecular epidemiol ogy, revealed that SC9 strains belonging to M. morganii subsp.morganii, possessed a higher number of ARGs compared to non-SC9 strains, mainly including genes conferring resistance to aminoglycosides, trimethoprims, sulfonamides, and tetracy clines, consistent with previous results (19).This discrepancy was particularly evident for aminoglycoside resistance genes.On average, SC9 strains contained 2.4 aminoglyco side resistance genes per strain, in contrast to 1.4 in non-SC9 strains.Specifically, a significantly higher prevalence of the aadA2, aph(3')-Ia, and ant(2'')-Ia genes was noted in SC9 strains.In addition, both this study and previous reports consistently demonstrate aminoglycoside resistance genes were frequently associated with MGEs (22), and SC9 strains harbor more MGEs compared to non-SC9 strains.Given the clinical preference for cephalosporins and aminoglycosides in treating M. morganii infections (6), we propose that the selection pressure from aminoglycosides and other antimicrobial agents along with the accumulation of MGEs could be one of the vital factors for the emergence and epidemic success of this sequence cluster.Furthermore, it is worth noting the near-uni versal presence of the hlyCABD gene cluster encoding the RTX toxin in SC9 strains, potentially enhancing their virulence compared to non-SC9 strains.Therefore, particular attention should be paid to the epidemiological development of the SC9 strains.In addition, for the first time, we discovered an association with the sequence cluster of M. morganii and its intrinsic resistance gene bla DHA type.The observation of similar phenomena has also been made within the Klebsiella oxytoca complex where sequence variations in the chromosomally-encoded β-lactamase gene, bla OXY , categorized the bacteria into different phylogroups (23).Considering M. morganii lacks MLST typing, the bla DHA gene may provide a foundational element for developing a gene-based typing system.
It is known that interplasmid horizontal transfer mediated by MGEs, such as transposons, plays a critical role in the emergence of novel plasmids.One illustrative example was the generation of conjugative plasmids simultaneously encoding for carbapenem resistance and hypervirulence.These plasmids contained a wide range of transposable elements that enabled them to undergo frequent genetic transposition (24).Kayoko Sugita elaborated on the interplasmid transposition of bla KPC-2 -containing Tn4401a from an IncN+R plasmid to a ColRNAI plasmid in Klebsiella pneumoniae (25).In this study, while the detection rate of the carbapenemase genes in M. morganii was relatively low, we found that most of these genes were harbored within plasmids.Notably, we also identified a novel plasmid carrying bla KPC-2 , which is inserted into M. morganii-specific plasmids by a transposon derived from E. coli.The identified plasmid, possessing conjugative transfer capability along with the carried ARGs, may contribute to the further dissemination of KPC-2-producing M. morganii.

Conclusions
In conclusion, our data indicate the significant contribution of MGEs in the evolutionary trajectory of two M. morganii subspecies.In M. morganii subsp.sibonii, typically isolated from environmental sources, the evolution of M. chanii genospecies has occurred, enhancing adaptation to the environment.In M. morganii subsp.morganii, mainly isolated from Homo sapiens, the SC9 strains, characterized by an increased abundance of ARGs and VRGs, might evolve into highly successful clones and pose a burgeoning challenge in healthcare settings.In addition, the novel bla KPC-2 -bearing plasmids, formed under the action of MGEs, may promote the spread of carbapenemase in M. morga nii.These observations underscore the critical importance of vigilant surveillance of M. morganii prevalence to prevent its emergence as a formidable obstacle in clinical therapeutics.

Bacterial isolates and identification
This study utilized a comprehensive surveillance program, selecting the clinical Morganella morganii isolates based on stringent criteria: the host had a clinical diagnosis indicating a potential infection, and only a single isolate per patient was included.Under this standard, a total of 206 M. morganii isolates were gathered from 45 hospitals across 26 provinces and municipalities that cover seven regions of China: northwestern China (n = 16), southern China (n = 25), northeastern China (n = 26), eastern China (n = 27), northern China (n = 35), central China (n = 38), and southwestern China (n = 39).The isolates were obtained from various specimens and hospital departments, as illustrated in Fig. 1A.Identification of all M. morganii isolates was performed using matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry.

Antimicrobial susceptibility testing
The bacterial antimicrobial susceptibility was tested using the broth microdilution method followed by Clinical and Laboratory Standards Institute (CLSI) standards (26).The susceptible breakpoints of cefoperazone defined by the CLSI were applied for cefoperazone-sulbactam.The susceptibilities of the 17 remaining antimicrobial agents were interpreted according to current CLSI guidelines (27).

Whole-genome data sequencing and genome analysis
The genomic DNA of all 206 isolates was subjected to draft-genome sequencing using a paired-end library on illumina novaseq 6000 system, and isolate 2387 was completegenome sequenced using PacBio RSII sequencer.After being removed low-quality sequences and adapters, the reads were de novo assembled by the SPAdes Genome Assembler (v3.11.1) and hybrid assembled using Unicycler (v0.4.6) (28,29).To enrich the analyzed data, we also retrieved a total of 225 reference genomes of M. morganii from NCBI genome data sets as of July 2023.Antibiotic resistome was characterized with abricate (v1.0.1) (30) screened against the ResFinder database (31).Due to the absence of virulence-related genes (VRGs) of M. morganii in the Virulence Factors of Pathogenic Bacteria Database (VFDB) (32), we established a list of virulence-related genes (the included virulence factors are listed in Table S1) for M. morganii based on the M. morganii KT strain (8) and the study by Palmieri et al. (13) and used BLASTx (v2.5.0) analysis for the detection of VRGs through this list (33).Plasmid types were identified using PlasmidFinder (v2.1) available at the Center for Genomic Epidemiology (CGE) (34).Comparisons of sequences from the novel bla KPC-2 -bearing plasmid discovered in this study were performed by BRIG (v0.95) (35), and Easyfig (v2.2.3) was used to visualize the linear alignment of the genetic structure of the transposition unit carrying bla KPC-2 (36).The set of mobile genetic elements (MGEs) including genomic islands (GIs), integrative conjugative elements (ICEs), integron, prophage, transposes (Tns), and insertion sequences (ISs) in all M. morganii were performed utilizing Mobilome Prediction at VRprofile2 database (37).

Conjugation experiments
The conjugation experiment was performed by the filter-mating method, employing the rifampin-resistant Escherichia coli EC600 as the recipient and bla KPC-2 -bearing isolate 2387 as the donor strain.In brief, isolate 2387 and the EC600 were separately cultured in Luria-Bertani (LB) broth at 37°C for 4 hours.Subsequently, the mixture was incubated on Mueller-Hinton (MH) agar plates with 1 µg/mL meropenem and 600 µg/mL rifampin to select the transconjugant.The conjugation frequency was calculated as the number of transconjugants divided by donors.All experiments were carried out three times.

Trehalose fermentation
Strains selected for this study included those carrying treR, treB, and treP, along with a negative control strain devoid of these genes.Bacterial suspensions were adjusted to a 0.5 McFarland standard and subsequently analyzed using the VITEK-2 system.

Statistical analysis
The Mann-Whitney U rank sum test was applied to pairwise group comparisons of ARGs, VRGs, and MGEs.Pearson's chi-square test or Fisher's exact test was applied to evalu ate the differences in antimicrobial resistance rates, ARGs, VRGs, and MGEs prevalence among the M. morganii.P values < 0.05 were considered statistically significant.All the statistical analyses were performed by Statistical Package for the Social Sciences (SPSS, v24.0).

FIG 2
FIG 2 The possible sources of formation for the novel bla KPC-2 -bearing plasmid in M. morganii.(A) The comparison between the novel bla KPC-2 -bearing plasmids and their most similar plasmids.The internal ring was the reference sequence of the novel bla KPC-2 -bearing plasmid p229813-KPC (MN310368.1).The outermost two rings were plasmids from M. morganii in the NCBI genome data sets, p11759-FII (MZ848139.1)and p46903_2 (CP070522.1),respectively.(B) Alignment of the genetic environment of transposition units carrying bla KPC-2 .On the left was the transposon unit from the novel plasmids p2387-KPC-2 and p229813-KPC, and on the right was the corresponding matching transposon unit from the plasmid unnamed3 in E. coli YL03 (CP093551).

FIG 3 FIG 4
FIG 3 The genomic classification of 431 M. morganii isolates.(A) Pairwise average nucleotide (ANI) identity comparison was calculated for all M. morganii isolates shown on a heatmap with blue indicating low and red indicating high nucleotide identity.The left diagram was an enlargement of a portion of the right diagram.(B) Principal component analysis (PCA) plot of the different genospecies of M. morganii.

FIG 5
FIG 5 The analysis of the antimicrobial resistance genes (ARGs), virulence-related genes (VRGs), and mobile genetic elements (MGEs) carried by different sequence clusters of M. morganii.(A) Variation in the quantity of ARGs, VRGs, and MGEs among different sequence clusters of M. morganii.(B) Profiles of ARGs, VRGs, and MGEs carried by different sequence clusters of M. morganii.* indicates that SC9 strains carry a significantly higher number of the corresponding category compared to non-SC9 strains (* represented a P-value of <0.05, and *** represented P < 0.01), and the circle represented non-SC9 strains carried more of the corresponding category.The box represents a high degree of dispersion in the number of resistance genes carried by different SCs (coefficient of variation greater than 15%).(C).Distribution of ARGs and VRGs with significant differences between SC9 and non-SC9 strains.

TABLE 2
The molecular epidemiology among the Morganella morganii collected from various host a Represented a P-value of <0.05.

TABLE 3
The molecular epidemiology between the two groups of Morganella morganii from NCBI genome data sets and this study a Represented a P-value of <0.05.