Letter to the Editor
Misidentification slanders Pantoea agglomerans as a serial killer
Madam,
Pantoea agglomerans is an ecologically versatile species that has important beneficial applications in agriculture as a biological alternative to antibiotic use for orchard protection against phytopathogenic bacteria.1 Disease-suppressive, biocontrol isolates are among the most effective tools avail- able to manage fire blight, an invasive threat to global apple and pear production. P. agglomerans-based products are registered for biocontrol in the USA, Canada and New Zealand, but clinical reports have been a regulatory obstacle in Europe.
Thus, the recent report of a fatal bacteraemia outbreak attributed to P. agglomerans is alarming and would have significant socio-economic impact, if the conclusion of clinical pathogenicity was correct.2
There is mounting evidence that many, if not most, P. agglomerans clinical reports are based upon inaccurate isolate identification, resulting from investigators’ reliance on inadequate identification methods and/or obsolete nomenclature.3 The P. agglomeranseEnterobacter agglom- eranseErwinia herbicola complex has experienced several taxonomic revisions,3 and both biochemical profiling as well as 16S rRNA gene sequencing lack the resolution needed to discriminate P. agglomerans sensu strictoisolates.4 Despite conclusive demonstration usinggyrBsequencing and matrix- assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) that a number of clinical strains archived in reputable culture collections do not belong to the genus Pantoea and even less to the species P. agglomerans, most remain erroneously catalogued.3,5 Inaccurate identification has been shown for 95.5% of
‘P. agglomerans’ clinical strains in the American Type Culture Collection (ATCC), a reference used to identify the Brazilian outbreak pathogen.3
The authors correctly recognized the difficulties involved in routine biochemical identification of P. agglomerans and appropriately appliedgyrB sequencing in their characteriza- tion of isolates from a nosocomial outbreak in Brazil.2 They reported their isolates as P. agglomerans based on a 100%
sequence identity of theirgyrBsequences with that of strain ATCC 27990, which was deposited in GenBank by our laboratory (GenBank nucleotide database: FJ617358). However, this strain is notP. agglomeransbut rather belongs to an unnamed
new species most closely related toEnterobacterspp. included within the Brenner’s Biotype XII.3,5 Nucleotide sequence identity between ATCC 27990 and type strain P. agglomerans LMG 1286T is 83% for gyrB, well below accepted species or genus thresholds (Figure 1). We have reconfirmed that a BLAST search of both the nucleotide and the protein databases using thegyrBsequence of ATCC 27990 returns not a singlePantoea within the 100 best hits, indicating a simple author error in database navigation and/or reporting. Regrettably, this error has resulted in an alarming misrepresentation of P. agglomeransas a ‘serial killer’. The potential for serious, lasting consequences on regulatory evaluation of plant- beneficial P. agglomerans biocontrol products necessitates correction of the clinical report and the related GenBank entries (nucleotide database: JN051137 and JN084037).
Conflict of interest statement None declared.
Funding sources
The Swiss Federal Office of Agriculture and the Swiss Commission for Technology and Innovation.
References
1. Pusey PL, Stockwell VO, Reardon CL,et al. Antibiosis activity of Pantoea agglomeransbiocontrol strain E325 againstErwinia amy- lovora on apple flower stigmas. Phytopathology 2011;101:
1234e1241.
2. Boszczowski I, No´brega de Almeida Ju´nior J, Peixoto de Miranda EJ, et al. Nosocomial outbreak ofPantoea agglomeransbacteraemia associated with contaminated anticoagulant citrate dextrose solu- tion: new name, old bug?J Hosp Infect2012;80:255e258.
3. Rezzonico F, Smits THM, Montesinos E,et al. Genotypic comparison ofPantoea agglomeransplant and clinical strains.BMC Microbiol 2009;9:204.
4. Rezzonico F, Stockwell VO, Tonolla M, et al. Pantoea clinical isolates cannot be accurately assigned to species based on meta- bolic profiling.Transpl Infect Dis2012;14:220e221.
5. Rezzonico F, Vogel G, Duffy B, et al. Application of whole-cell matrix-assisted laser desorption ionizationetime of flight mass spectrometry for rapid identification and clustering analysis of Pantoeaspecies.Appl Environ Microbiol2010;76:4497e4509.
Available online atwww.sciencedirect.com
Journal of Hospital Infection
j o u r n a l h o m e p a g e : w w w . e l s e v i e r h e a l t h . c o m / j o u r n a l s / j h i n
0195-6701/$esee front matterÓ2012 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jhin.2012.02.013
Journal of Hospital Infection 81 (2012) 137e139
Pantoea agglomerans LMG 1286 CIP A181
ATCC 27987
Pantoea eucalypti LMG 24198 Pantoea conspicua LMG 24534
Pantoea brenneri LMG 5343 (=ATCC 29921) ATCC 29001
Pantoea anthophila LMG 2558 (=NCPPB 1682**) Pantoea vagans C9-1
Pantoea deleyi LMG 24200
Pantoea stewartii pv. indologenes CFBP 3517 Pantoea stewartii pv. stewartii CFBP 3614
Pantoea allii LMG 24202 ATCC 27995
Pantoea ananatis LMG 5342 ATCC 27996
Pantoea septica LMG 5345 (=ATCC 29923) Pantoea sp. nov. EM486
Pantoea toletana LMG 24162 Pantoea calida 1400-07
Erwinia papayae DSM 16540 Erwinia mallotivora LMG 1270
Erwinia psidii BCC 1322 Erwinia oleae CFBP 6632
Erwinia billingiae Eb661 Erwinia pyrifoliae DSM 12163
Erwinia tasmaniensis Et1/99 Erwinia piriflorinigrans CFBP 5888 Erwinia amylovora CFBP 1232
Erwinia persicina CFBP 3622 ATCC 27991
ATCC 27992
Erwinia aphidicola JCM 21239 LMG 5341* (=ATCC 29919)
Pantoea rodasii LMG 26273 Pantoea rwandensis LMG 26276
Pantoea eucrina LMG 5346 (=ATCC 27998) 100
75 100 100 100
100
98 72
100 99 99
98
100 99 78
99
55
97 71
95
53
54
Pantoea wallisii LMG 26277 Pantoea dispersa LMG 2603 CIP 102701
Pantoea cypripedii LMG 2657
Dickeya dadantii LMG 2804
Pectobacterium carotovorum SCRI1043 Brenneria nigrifluens LMG 2694 Enterobacter sakazakii ATCC 51329
ATCC 27985 ATCC 27993 ATCC 27994
LMG 5336* (=ATCC 29904)
Enterobacter cloacae ATCC 13047 Enterobacter cancerogenus LMG 2693 Brenneria rubrifaciens LMG 2709
CIP 102343
Klebsiella pneumoniae 342
Salmonella enterica sv. Typhi CT18 CECT 4842
ATCC 27988 LMG 5340* (=ATCC 29918) Citrobacter koseri ATCC BAA-895
Escherichia coli K-12 MG1655 ATCC 27982
ATCC 27989 LMG 5337* (=ATCC 27915) ATCC 27981
ATCC 27990
LMG 5339* (=ATCC 27917)
Tatumella terrea LMG 22051 Tatumella morbirosei LMG 23359
Tatumella punctata LMG 22097 100
100 100
95 99 100
100
100 78
97
89 95
66 73 54
0.02
Figure 1. Phylogenetic position of ATCC 27990 (yellow box) used as the reference to identify a Brazilian outbreak pathogen, compared with truePantoea agglomerans(grey box) and clinical isolates received from culture collections asP. agglomeransor basonymsErwinia herbicolaandEnterobacter agglomerans.
F. Rezzonico*
T.H.M. Smits B. Duffy Agroscope Changins-Wa¨denswil ACW, Plant Protection Division, Research Bacteriology, Wa¨denswil, Switzerland
* Corresponding author. Address: Schloss 1, P.O. Box, CH-8820 Wa¨denswil, Switzerland. Tel.:þ41 44 783 6111;
fax:þ41 44 780 6341.
E-mail address:fabio.rezzonico@acw.admin.ch (F. Rezzonico).
Available online 1 May 2012 Letter to the Editor / Journal of Hospital Infection 81 (2012) 137e139 139
