LETTER
Shortly after the discovery of Elizabethkingia anophelis from the mosquito Anopheles gambiae, clinically significant infections attributed to this species were repeatedly described (1); they include a large outbreak involving 69 isolates in Wisconsin, USA (2), and also an intensive-care unit outbreak that occurred at our hospital (3). E. anophelis is genetically distinct from Elizabethkingia meningoseptica and Elizabethkingia miricola (4, 5). However, factory default databases accompanying the matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) identification systems, such as Vitek mass spectroscopy (MS) (bioMérieux, Marcy l'Etoile, France) and the Bruker MALDI Biotyper (Bruker Daltonics, Bremen, Germany), cannot accurately distinguish between E. anophelis and E. meningoseptica unless the user supplies his/her own mass spectrometry data set of characterized isolates (1, 6, 7).
In this study, we were interested to see if there were misidentified cases of E. anophelis among our Elizabethkingia spp. Seventy-nine retrospective blood isolates spanning the time period from 2009 to May 2017 were analyzed. All isolates were identified as either E. meningoseptica (96.2%) or E. miricola (3.8%) by the Bruker MALDI Biotyper (bioMérieux). Almost-full-length 16S rRNA gene sequencing was performed using the universal primers fD1 (5′-AGAGTTTGATCCTGGCTCAG-3′) and rP2 (5′-ACGGCTACCTTGTTACGACTT-3′), with sequencing results indicating that 78/79 (98.7%) isolates were E. anophelis and that the remaining single isolate was E. meningoseptica. Specifically, 73 isolates had 99.8 to 100% nucleotide identity to E. anophelis strain 0422, 2 isolates had 100% nucleotide identity to E. anophelis strain 107618, and 1 isolate each had 100% nucleotide identity to E. anophelis strain HKU36, E. anophelis strain F3543, or E. endophytica strain F3201. E. endophytica has since been recognized as an additional strain of E. anophelis and not as a separate species (8). These results reiterate the genomic heterogeneity of this species (5). The E. meningoseptica isolate had 100% 16S sequence identity to E. meningoseptica strain KC1913. Strains NUHP1 (GenBank accession number NZ_CP007547.1) and E. meningoseptica ATCC 13253 (GenBank accession number ASAN00000000.1) were used as controls for both the sequencing and the PCR validation assay.
For rapid PCR differentiation between E. anophelis and E. meningoseptica, species-specific gene targets were identified based on the comparative analysis of the complete genomes (as of 31 August 2017) of E. anophelis, E. meningoseptica, and E. miricola using the MAUVE software (9). The gene encoding lipid A-disaccharide synthase (GenBank locus tag BD94_RS01570) was utilized for E. anophelis detection with primers anoR (5′-TGCGTTATTACCAGGTAGTCGG-3′) and anoF (5′-GACTTCCGCGGTAGCAAACAA-3′). For the detection of E. meningoseptica, a putative sodium-proton antiporter (GenBank locus tag BBD35_RS10505) was targeted, using primers mengF (5′-TGGGACCTATTGCTGTTGGTT-3′) and mengR (5′-ACCACTTCCTGTGTACCTGC-3′). PCR amplifications were performed using a Taq PCR master mix kit (Qiagen, Hilden, Germany), with an initial denaturation at 95°C for 5 min, followed by 35 temperature cycles of heat denaturation at 94°C for 30 s, primer annealing at 55°C for 30 s, and extension at 72°C for 30 s, with a final extension step at 72°C for 10 min. Agarose gel electrophoresis was used for analysis of PCR products. Amplicon sizes for E. anophelis and E. meningoseptica detection were 281 bp and 250 bp, respectively. The analytical specificities of these 2 primer sets were determined using 81 Elizabethkingia blood isolates as well as 42 clinical isolates of Gram-negative bacteria (Table 1). The primers demonstrated their intended specificity, and no PCR cross-reactivity was observed.
TABLE 1.
Elizabethkingia spp. and other Gram-negative bacteria used for evaluating the analytical specificities of E. anophelis and E. meningoseptica primersa
| Bacterium | No. of isolates | PCR result with: |
|
|---|---|---|---|
| E. anophelis-specific (anoR and anoF) primers | E. meningoseptica-specific (mengF and mengR) primers | ||
| E. anophelis | 79 | + | — |
| E. meningoseptica | 2 | — | + |
| Achromobacter xylosoxidans | 3 | — | — |
| Chryseobacterium spp. | 2 | — | — |
| Stenotrophomonas maltophilia | 3 | — | — |
| Burkholderia gladioli | 1 | — | — |
| Burkholderia cepacia complex | 4 | — | — |
| Pseudomonas aeruginosa | 3 | — | — |
| Pseudomonas putida | 2 | — | — |
| Acinetobacter baumannii | 4 | — | — |
| Escherichia coli | 4 | — | — |
| Enterobacter cloacae complex | 4 | — | — |
| Enterobacter aerogenes | 4 | — | — |
| Klebsiella pneumoniae | 5 | — | — |
| Klebsiella oxytoca | 3 | — | — |
The Elizabethkingia spp. were determined to the species level based on 16S rRNA gene sequencing, while the other non-Elizabethkingia isolates were identified using MALDI-TOF mass spectrometry. +, PCR positive; —, no amplicons were detected.
Susceptibility testing was performed using broth microdilution Sensititre GNX3F plates (Thermo Fisher Scientific, MA, USA). The isolates were resistant to multiple drug classes, including aminoglycosides, third-generation cephalosporins, and carbapenems (Table 2). This correlated with the genomic content of E. anophelis, which harbors multiple antimicrobial resistance genes as well as metallo-β-lactamases (5). However, the majority of the isolates remained susceptible to levofloxacin and trimethoprim-sulfamethoxazole (Table 2). This was in contrast to the results of a Korean study in which 65% and 78% of E. anophelis isolates were resistant to levofloxacin and trimethoprim-sulfamethoxazole, respectively (7). A Hong Kong study reported that 70.6% of isolates were susceptible to trimethoprim-sulfamethoxazole (1). The Korean study employed an agar dilution method for susceptibility testing (7), while the Hong Kong group used the Vitek 2 GNI system (1). The differences observed in the MICs may have resulted from the different testing methods used, although geographical variation of isolates cannot be ruled out.
TABLE 2.
Susceptibility profiles for 79 Elizabethkingia species strains isolated from blood
| Antibiotic(s)a | MIC range (mg/liter) | MIC(s) (mg/liter) |
Susceptibility (%)b |
|||
|---|---|---|---|---|---|---|
| MIC50 | MIC90 | S | I | R | ||
| Amikacin | 4 to >32 | >32 | >32 | 6.3 | 35.4 | 58.2 |
| Gentamicin | 4 to >8 | >8 | >8 | 1.3 | 3.8 | 94.9 |
| Tobramycin | 8 to >8 | >8 | >8 | 0.0 | 3.8 | 96.2 |
| Levofloxacin | ≤1 to >8 | 2 | >8 | 78.5 | 2.5 | 19.0 |
| Ciprofloxacin | 0.25 to >2 | >2 | >2 | 21.5 | 26.6 | 51.9 |
| Colistin | >4 | >4 | >4 | 0.0 | 0.0 | 100.0 |
| Polymyxin B | >4 | >4 | >4 | 0.0 | 0.0 | 100.0 |
| Doxycycline | <2 to 8 | <2 | 4 | 92.4 | 7.6 | 0.0 |
| Minocycline | <2 to >16 | <2 | <2 | 97.5 | 1.3 | 1.3 |
| Tigecycline | <0.25 to >8 | 2 | 8 | 5.1 | 1.3 | 92.4 |
| Trimethoprim-sulfamethoxazole | 0.5 to 4, 9.5 to 76 | 1, 19 | 2, 38 | 92.4 | 0.0 | 7.6 |
| Ceftazidime | >16 | >16 | >16 | 0.0 | 0.0 | 100.0 |
| Cefotaxime | 16 to >32 | 32 | >32 | 0.0 | 86.1 | 13.9 |
| Cefepime | 16 to >16 | >16 | >16 | 0.0 | 36.7 | 63.3 |
| Imipenem | 8 to >8 | >8 | >8 | 0.0 | 2.5 | 97.5 |
| Meropenem | 8 to >8 | >8 | >8 | 0.0 | 3.8 | 94.9 |
| Doripenem | >4 | >4 | >4 | 0.0 | 0.0 | 100.0 |
| Aztreonam | 8 to >16 | >16 | >16 | 1.3 | 2.5 | 96.2 |
| Piperacillin-tazobactam | 8 to 64 | 16 | 16 | 92.4 | 6.3 | 1.3 |
| Ticarcillin-clavulanic acid | 16 to >128 | 64 | >128 | 21.5 | 27.8 | 50.6 |
| Ampicillin-sulbactam | 16 to >64 | 64 | 64 | 0.0 | 0.0 | 100.0 |
For drug combinations, tazobactam was held constant at 4 mg/liter, clavulanic acid was held constant at 2 mg/liter, the ampicillin-sulbactam combination was in a 2:1 ratio, and the trimethoprim-sulfamethoxazole combination was in a 1:19 ratio.
S, susceptible; I, intermediate; R, resistant. The interpretive criteria applied were those of the Clinical and Laboratory Standards Institute (CLSI) for non-Enterobacteriaceae, except for tigecycline, doripenem, colistin, and polymyxin B, for which CLSI breakpoints were unavailable. For tigecycline and doripenem, the European Committee on Antimicrobial Susceptibility Testing (EUCAST) non-species-specific breakpoints were used. For colistin and polymyxin B, Pseudomonas aeruginosa CLSI breakpoints were used.
In summary, we describe the overwhelming predominance of E. anophelis from blood isolates, suggesting that this is not a rare pathogen, and in actual fact, E. meningoseptica may be the more uncommon species. The development of a species-specific PCR should prove useful for the differentiation of E. anophelis and E. meningoseptica, particularly for routine laboratories relying on either MALDI-TOF MS (without an extended database) for identification.
REFERENCES
- 1.Lau SKP, Chow WN, Foo CH, Curreem SO, Lo GC, Teng JL, Chen JH, Ng RH, Wu AK, Cheung IY, Chau SK, Lung DC, Lee RA, Tse CW, Fung KS, Que TL, Woo PC. 2016. Elizabethkingia anophelis bacteremia is associated with clinically significant infections and high mortality. Sci Rep 6:26045. doi: 10.1038/srep26045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Perrin A, Larsonneur E, Nicholson AC, Edwards DJ, Gundlach KM, Whitney AM, Gulvik CA, Bell ME, Rendueles O, Cury J, Hugon P, Clermont D, Enouf V, Loparev V, Juieng P, Monson T, Warshauer D, Elbadawi LI, Walters M, Crist MB, Noble-Wang J, Borlaug G, Rocha EPC, Criscuolo A, Touchon M, Davis JP, Holt KE, McQuiston JR, Brisse S. 2017. Evolutionary dynamics and genomic features of the Elizabethkingia anophelis 2015 to 2016 Wisconsin outbreak strain. Nat Commun 8:15483. doi: 10.1038/ncomms15483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Teo J, Tan SY-Y, Tay M, Ding Y, Kjelleberg S, Givskov M, Lin RT, Yang L. 2013. First case of E anophelis outbreak in an intensive-care unit. Lancet 382:855–856. doi: 10.1016/S0140-6736(13)61858-9. [DOI] [PubMed] [Google Scholar]
- 4.Kämpfer P, Matthews H, Glaeser SP, Martin K, Lodders N, Faye I. 2011. Elizabethkingia anophelis sp. nov., isolated from the midgut of the mosquito Anopheles gambiae. Int J Syst Evol Microbiol 61:2670–2675. doi: 10.1099/ijs.0.026393-0. [DOI] [PubMed] [Google Scholar]
- 5.Breurec S, Criscuolo A, Diancourt L, Rendueles O, Vandenbogaert M, Passet V, Caro V, Rocha EP, Touchon M, Brisse S. 2016. Genomic epidemiology and global diversity of the emerging bacterial pathogen Elizabethkingia anophelis. Sci Rep 6:30379. doi: 10.1038/srep30379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Janda JM, Lopez DL. 2017. New pathogen profiles: Elizabethkingia anophelis. Diagn Microbiol Infect Dis 88:201–205. doi: 10.1016/j.diagmicrobio.2017.03.007. [DOI] [PubMed] [Google Scholar]
- 7.Han MS, Kim H, Lee Y, Kim M, Ku NS, Choi JY, Yong D, Jeong SH, Lee K, Chong Y. 2017. Relative prevalence and antimicrobial susceptibility of clinical isolates of Elizabethkingia species based on 16S rRNA gene sequencing. J Clin Microbiol 55:274–280. doi: 10.1128/JCM.01637-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Doijad S, Ghosh H, Glaeser S, Kämpfer P, Chakraborty T. 2016. Taxonomic reassessment of the genus Elizabethkingia using whole-genome sequencing: Elizabethkingia endophytica Kämpfer et al. 2015 is a later subjective synonym of Elizabethkingia anophelis Kämpfer et al. 2011. Int J Syst Evol Microbiol 66:4555–4559. doi: 10.1099/ijsem.0.001390. [DOI] [PubMed] [Google Scholar]
- 9.Darling AE, Mau B, Perna NT. 2010. progressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One 5:e11147. doi: 10.1371/journal.pone.0011147. [DOI] [PMC free article] [PubMed] [Google Scholar]