INTRODUCTION
Gram-negative bacterium
Elizabethkingia meningoseptica, which is non-motile and exhibits catalase and oxidase positivity, as well as non-glucose fermentation, is commonly found in both natural environments and hospital settings, including water, soil, plants, foodstuffs, and medical devices (
1 – 3). Initially known as
Flavobacterium meningosepticum,
E. meningoseptica was first identified in 1959 by Elizabeth O. King (
4). It was later renamed
C. meningosepticum in 1994 when it was classified into a new genus,
Chryseobacterium (
2). However, in 2005, through analysis of the 16S rRNA gene sequence and phylogenetic tree, it was reassigned to a new genus,
Elizabethkingia (
1). While healthy individuals rarely contract
E. meningoseptica infections or diseases, there has been an increasing number of reports linking this bacterium to life-threatening infections in immunocompromised individuals. It has been associated with severe meningitis (particularly neonatal meningitis) (
5 – 11), bacteremia (
12 – 17), respiratory infection (
18,
19), urinary tract infection (
20), sepsis (
21,
22), eye infections (
23 – 26), biliary tract infections (
27,
28) and has emerged as a significant public health concern. Furthermore, recent studies have described nosocomial outbreaks associated with
E. meningoseptica (
5,
6,
21,
29 – 31). It is also concerning that previous research has shown a mortality rate of up to 40% with
E. meningoseptica infections, especially in neonates (
14,
15,
32). For example, among 19 mechanically ventilated patients in acute care hospitals affected by an outbreak, 8 ultimately died (
5).
Treatment of
E. meningoseptica infection is challenging due to the lack of effective treatment options and this microorganism’s reduced sensitivity to many classes of antimicrobials. The organism is generally resistant to antimicrobials that are effective against Gram-negative bacteria, such as aminoglycosides, chloramphenicol, extended-spectrum beta-lactams, carbapenems, colistin, and even vancomycin (
12,
14,
15,
33 – 37). Limited information is available regarding
E. meningoseptica’s pathogenesis, resistance mechanisms, and direct transmission routes (
38). In-depth genome analyses will provide insights into the evolutionary history, transmission pathways, pathogenesis, and resistance mechanisms of
E. meningoseptica. Whole genome analysis can offer molecular diagnostic tools, such as single nucleotide polymorphisms (SNPs), which have potential clinical utility.
E. meningoseptica can benefit from whole genome analysis as it is a relatively understudied bacterium with few genomes available in the NCBI database. A review of the existing literature reveals that almost all publications on this microorganism are case reports, with only a few studies investigating its comparative genomics, such as phylogenetic structure and geographical distribution. Here, we have isolated over 20 clinical
E. meningoseptica strains and sequenced their complete genomes. Various methods were utilized to analyze and compare the genome characteristics of all GenBank sequences of
E. meningoseptica bacteria worldwide. This collection represents the largest and most geographically diverse sample to date. Our study enables an investigation into the population structure, evolutionary history, geographical distribution, transmission assessment, virulence, and resistance mechanisms of the studied group.
DISCUSSION
E. meningoseptica is ubiquitously distributed in nature, including soil, water, and hospitals. Recently, it has been increasingly recognized as a pathogen that can cause nosocomial infections in immunocompromised individuals. In this study, we collected over 20 clinical isolates of E. meningoseptica around a 10-year period and conducted complete genome sequencing and comparative genomics analysis. Whole genome sequencing and analysis have become important tools in studying pathogens, thanks to advancements in molecular biology and biotechnology. Previous research has mainly focused on case studies and the susceptibility of E. meningoseptica to antibacterial agents. However, there is limited available genomic information on this bacterium. To the best of our knowledge, this is the first comprehensive analysis of the evolutionary relationships and comparisons among E. meningoseptica bacteria using extensive genome sequencing.
According to a previous study, diagnosing uncommon non-fermenting bacteria poses difficulties (
40,
41). The current methods for distinguishing between
Chryseobacterium indologenes and
E. meningoseptica lack reliability and consensus (
40,
41). Traditional methods are challenging for identifying
C. indologenes and
E. meningoseptica. Previous studies have shown that these techniques are not highly effective in distinguishing between
Chryseobacterium and
Elizabethkingia species (
40 – 43). Among automated phenotypic methods, VITEK 2, MALDI-TOF MS, MALDI-BD (MALDI-TOF BioTyper), Phoenix 100 ID/AST, and API (Analytical Profile Index) are used to identify
E. meningoseptica isolates. However, due to limited database coverage, other
Elizabethkingia species are often misidentified as
E. meningoseptica (
40,
42,
43). Especially, the API/ID 32 v3.1 system has an identification rate of less than 30% for
Elizabethkingia species (
42). Recent papers have also shown that
Elizabethkingia anophelis is frequently misidentified as
E. meningoseptica using current commercial identification systems (
43 – 45). Therefore, it is recommended to modify the method for discriminating
E. meningoseptica from
Chryseobacterium gleum and other
Elizabethkingia species and to expand the database coverage for
E. anophelis in the discussed microbial identification systems. Generally speaking, 16S rRNA gene sequencing is the preferred method for strain identification. While this traditional approach can successfully identify most novel strains, it may encounter inaccuracies when dealing with rare strains. Our investigation revealed the presence of 35 additional
E. meningoseptica genomes, including raw data in Sequence Read Archive (SRA) in the NCBI database, apart from the strains we obtained. After conducting an ANI analysis, we found that the values of 13 isolates were significantly low (less than 80%) to be considered as
E. meningoseptica or even belonging to the same genus (Fig. S7). Interestingly, the ANI value of isolate 5453STDY7605978 among these 13 strains was relatively low, at 62.73%, indicating a potential error in either the sequencing or submission process of the genome (Fig. S7). To address these issues, it is recommended to design new specific primers for species identification through 16S rRNA gene sequencing. These findings emphasize the importance of obtaining ANI values prior to conducting a genome analysis of rare and novel pathogenic bacteria.
Several previous studies have shown that
Elizabethkingia species isolated from different geographic regions are susceptible to different antibiotics and exhibit complex antimicrobial resistance characteristics (
15,
45). Our study results indicate that the geographic distribution of
E. meningoseptica resistance genes is not linked. It is intriguing to consider how the biofilm may interpret this phenomenon and the multidrug resistance mechanism of
E. meningoseptica species. A recent study demonstrated that
E. meningoseptica bacteria have the ability to form biofilms and adhere (
46). Our investigation revealed a great number of genes associated with biofilm formation. Biofilm has become a prominent topic of discussion in the context of antibiotic resistance in pathogenic bacteria in recent years. Chloramphenicol is not typically used to treat internal bacterial infections nor has it been investigated for
Elizabethkingia species. Remarkably, chloramphenicol has been found to be a potent antibiotic capable of destroying various bacteria that can cross the blood-brain barrier, and it also possesses anti-biofilm properties (
47). Chloramphenicol could potentially be utilized to treat meningitis infections caused by
E. meningoseptica.
We hypothesized that many
E. meningoseptica isolates were connected through transmission networks, as they were less than 10 SNPs apart from one another, regardless of the origin, whether it was a person or a different nation. This situation is particularly interesting. This discovery is of great importance as it implies that
E. meningoseptica could be transmitted within a hospital, across borders, and even cause notorious outbreaks of infection. It has been challenging to find epidemiological evidence supporting the existence of direct genetic super-spreaders. Since the mode of transmission of
E. meningoseptica is unclear and the organism is resistant to multiple classes of antimicrobials, treating an infection caused by this organism is a daunting task. This typical pattern of antimicrobial susceptibility can hinder the selection of the most suitable medications. In contrast, there is a scarcity of antimicrobial data for
E. meningoseptica, and the results of susceptibility testing can vary depending on the method used (
14). Given the difficulty in treating this organism with antimicrobials, it is advisable to use susceptibility testing to guide the choice of treatment. It has been suggested that patients who require long-term acute care with mechanical ventilation may be a significant source of transmission for the multi-drug resistant pathogen,
E. meningoseptica, following the outbreak description (
5).
There are several limitations to our present study. Firstly, it is a retrospective and single-center investigation, as our
E. meningoseptica strain was predominantly obtained from a tertiary medical center. Secondly, the sample size of this study was limited, despite the inclusion of samples from 2010 to 2019, and the sources were diverse, which may not effectively represent the broader distribution in China. To address these limitations, it is necessary to sequence a larger number of
E. meningoseptica strains from China and other regions worldwide. Furthermore, it is important to note that no minimum inhibitory concentration (MIC) cut-off values have been established by the Clinical and Laboratory Standards Institute (CLSI) and European Committee on Antimicrobial Susceptibility Testing for
E. meningoseptica strains. Therefore, this article relies on published literature to develop interpretive criteria for MIC (
48,
49).
In summary, E. meningoseptica has recently been identified as a pathogen that can cause severe and potentially lethal infections in humans. Variations in resistance and virulence genes further support the existing evidence of the natural resistance and pathogenic nature of E. meningoseptica. Our research emphasizes the possibility of a significant outbreak of E. meningoseptica in hospitals and its potential to spread internationally. It is recommended that increased genomic monitoring be carried out to gain a deeper understanding of the dynamics of E. meningoseptica populations.