Title: Outbreak of a novel Enterobacter sp. carrying
blaCTX−M−15in a neonatal unit of a tertiary hospital in
Authors: Stephen E. Mshana, Lisa Gerwing, Mercy Minde,
Torsten Hain, Eugen Domann, Eligius Lyamuya, Trinad
Chakraborty, Can Imirzalioglu
To appear in:
Please cite this article as: Mshana SE, Gerwing L, Minde M, Hain T, Domann E,
Lyamuya E, Chakraborty T, Imirzalioglu C, Outbreak of a novel Enterobacter sp.
carrying blaCTX−M−15in a neonatal unit of a tertiary hospital in Tanzania, International
Journal of Antimicrobial Agents (2010), doi:10.1016/j.ijantimicag.2011.05.009
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Page 1 of 23
Outbreak of a novel Enterobacter sp. carrying blaCTX-M-15 in a neonatal
unit of a tertiary hospital in Tanzania
Stephen E. Mshana a, Lisa Gerwing a, Mercy Minde a, Torsten Hain b, Eugen Domann b,
Eligius Lyamuya c, Trinad Chakraborty b,*, Can Imirzalioglu b,*
a Weill Bugando University College of Health Sciences, P.O. Box 1464, Mwanza,
b Institute of Medical Microbiology, Justus-Liebig-University of Giessen, Frankfurter
Strasse 107, D-35392 Giessen, Germany
c Muhimbili University of Health and Allied Sciences, Box 65015, Dar es Salaam,
Received 31 January 2011
Accepted 9 May 2011
Novel Enterobacter sp.
Page 2 of 23
* Corresponding authors. Tel.: +49 641 994 1251; fax: +49 641 994 1259.
E-mail addresses: Can.Imirzalioglu@mikrobio.med.uni-giessen.de (C. Imirzalioglu);
Trinad.Chakraborty@mikrobio.med.uni-giessen.de (T. Chakraborty).
Page 3 of 23
multiresistant Enterobacter strain in an African neonatal unit that can easily be
Enterobacter hormaechei and Cronobacter sakazakii are among the most important
causes of outbreaks of neonatal sepsis associated with powdered milk. In this study, we
report for the first time an outbreak of a novel Enterobacter sp. harbouring blaCTX-M-15 in
a neonatal unit in Tanzania. Seventeen Gram-negative enteric isolates from neonatal
blood cultures were studied. Antibiotic susceptibility was assessed by disk diffusion
testing, and the presence of the blaCTX-M-15 gene was established by polymerase chain
reaction (PCR) and sequencing. Isolates were typed by pulsed-field gel electrophoresis
(PFGE). Identification by biochemical profiling was followed by nucleotide sequencing of
16S ribosomal DNA (rDNA), rpoB and hsp60 alleles. Environmental sampling was done
and control measures were established. Isolates were initially misidentified based on
their fermentation characteristics and agglutination as Salmonella enterica serotype
Paratyphi. All isolates were resistant to multiple antibiotics, except for ciprofloxacin and
carbapenems, and were found to harbour blaCTX-M-15 on a 291-kb narrow-range plasmid.
PFGE analysis indicated the clonal outbreak of a single strain, infecting 17 neonates
with a case fatality rate of 35%. The same strain was isolated from a milk bucket.
Phylogenetic analysis using 16S rDNA, rpoB and hsp60 sequences permitted no
definitive identification, clustering the strains in the Enterobacter cloacae complex with
similarities of 92–98.8%. The data describe an outbreak of a novel blaCTX-M-15-positive,
misidentified taxonomically. These data highlight the need for constant surveillance of
bacteria harbouring extended-spectrum -lactamases as well as improvements in
hygiene measures in developing countries.
Page 4 of 23
Enterobacter spp., especially Cronobacter (Enterobacter) sakazakii, have been
associated with neonatal sepsis through ingestion of contaminated powdered milk . A
number of outbreaks of C. sakazakii have been reported in developed countries [2,3],
whereas information from developing countries is limited. Bacteria of the genus are
ubiquitous in nature and can cause severe infection, especially in immunocompromised
neonates . Members of the Enterobacter cloacae complex are most frequently
isolated from clinical samples. In this complex, Enterobacter hormaechei isolates appear
to be prone to carrying extended spectrum -lactamase (ESBL) genes  and can be
phenotypically misidentified as C. sakazakii when using commercially available
biochemical kits . Identification of Enterobacter spp. is routinely done by phenotypic
methods. Most clinical laboratories in developed countries employ commercially
available kits or semiautomated systems, whilst in developing countries simple in-house
biochemical panels are commonly applied . For further identification and
discrimination of the species in this genus, 16S ribosomal RNA (rRNA), rpoB and hsp60
gene sequencing have been used. Here we report the outbreak of a novel species of
Enterobacter carrying blaCTX-M-15, which can easily be misidentified using routinely
applied biochemical kits [2,4,6].
Page 5 of 23
Susceptibility patterns to ampicillin, amoxicillin/clavulanic acid (AMC),
2.1. Bacterial isolates
Bacterial strains were isolated from blood cultures of neonates from the Bugando
Medical Centre (Mwanza, Tanzania) between December 2009 and February 2010.
Environmental sampling was performed, with samples obtained of solutions, milk,
suctions tubes, sinks and mattresses.
Bacteria were identified using in-house biochemical and fermentation assays (Triple
sugar iron agar, citrate utilisation, indole production, urease test, hydrogen sulphide
production, motility testing). A Salmonella polyvalent latex agglutination test (Oxoid Ltd.,
Basingstoke, UK) was performed for all isolates. Further identification was done using
API 20E (bioMérieux, Marcy l'Étoile, France), VITEK 2 GN ID Card (bioMérieux) and
Phoenix-NMIC/ID-64 (Becton Dickinson, Heidelberg, Germany). Fatty acid analysis was
performed by the German Resource Centre for Biological Materials (DSMZ). In addition,
strains were examined for their ability to form biofilms on plastic surfaces as described
by O’Toole et al. .
sulfamethoxazole/trimethoprim (SXT), tetracycline, gentamicin, ciprofloxacin,
chloramphenicol, fosfomycin, ceftriaxone, cefotaxime, cefepime, ceftazidime and
meropenem were determined by disk diffusion following Clinical and Laboratory
Page 6 of 23
CC118 (RifR, StrR, Lac–, plasmid-free) as a recipient strain. Transconjugants were
Standards Institute (CLSI) guidelines . Escherichia coli ATCC 25922 was used as the
quality control strain. ESBL production was confirmed using disk approximation methods
Pulsed-field gel electrophoresis (PFGE) analysis was performed as previously described
(http://www.cdc.gov/pulsenet/protocols.htm). Briefly, genomic DNA was prepared in
agarose blocks and was digested with the restriction enzyme XbaI (Fermentas, St Leon-
Rot, Germany). DNA products were separated by PFGE on a CHEF-DRIVE III
apparatus (Bio-Rad, La Jolla, CA) in 0.5 Tris–borate–ethylene diamine tetra-acetic acid
(TBE) buffer for 20 h at 6 V/cm at 13 C, 5–50 s for 20 h. A dendrogram illustrating the
genetic similarity of 18 strains was generated from banding patterns using GelCompar
II® (Applied Maths, Sint-Martens-Latem, Belgium). Strains showing ≥80% similarity were
classified as genetically related and were assigned to the same clusters.
The presence of ESBLs was established by polymerase chain reaction (PCR) with
specific primers targeting SHV, TEM and CTX-M genes . Sequencing of the PCR
products was done to establish the corresponding ESBL allele (LGC Genomics GmbH,
Berlin, Germany). To determine the location of ESBL genes, conjugation experiments
were carried out as described previously  with two isolates as donors and E. coli
selected on lysogeny broth (LB) plates supplemented with 30 g/mL cefotaxime and 300
g/mL rifampicin and were confirmed for the presence of ESBL genes using phenotypic
and molecular methods. In situ digestion of agarose blocks with S1 nuclease was used
Page 7 of 23
of hydrogen sulphide. All strains gave a positive agglutination test using a polyvalent
to determine plasmid size [10,11] as described previously, followed by Southern blotting
and hybridisation using digoxigenin (DIG)-labelled blaCTX-M-15 amplicon probes, prepared
according to the manufacturers instructions (DIG High Prime DNA Labelling and
Detection Starter Kit II; Roche, Mannheim, Germany).
A 1444-bp 16S rDNA template was amplified and sequenced using the primers 49F
(TWAYACATGCAAGTCGRRCG), 1504R (CTTGTTACGACTTCACCCCAG), 355R
(GCTGCCTCCCGTAGCAGTCTGG) and 1092F (AAGTCCCGCAACGAGCGCAAC),
followed by rpoB and hsp60 phylogeny studies .
3. Results and discussion
During December 2009 to February 2010, a Gram-negative bacterium with identical
antibiotic profile was isolated from blood samples of 17 neonates (Table 1). A similar
strain was isolated from the milk bucket. The first case came from the neonatal Intensive
Care Unit (NICU) and was followed by more cases in the NICU as well as the general
neonatal unit. Twelve cases (71%) were admitted in January 2010 (Table 1). Using in-
house phenotypic biochemical profiling, the isolates were classified as non-lactose-
fermenters, motile, indole-negative, citrate-positive, urease-negative with no production
Salmonella kit; thus, the identification of Salmonella enterica serotype Paratyphi was
Page 8 of 23
All neonates were initially treated with ampicillin, gentamicin and cefotaxime without
By disk diffusion test, all strains were found to be resistant to ampicillin [30 g; mean ±
standard deviation (S.D.) zone diameter 6.06 ± 0.2356 mm], AMC (20/10 g; mean ±
S.D. zone diameter 11.78 ± 0.428 mm), gentamicin (10 g; mean ± S.D. zone diameter
6.22 ± 0.428 mm), SXT (25/23.75 g; mean ± S.D. zone diameter 6.56 ± 0.856 mm),
tetracycline (30 g; mean ± S.D. zone diameter 6 ± 0.000 mm), fosfomycin (200 g;
mean ± S.D. zone diameter 8.67 ± 0.485 mm), chloramphenicol (30 g; mean ± S.D.
zone diameter 10.78 ± 0.428 mm), cefotaxime (30 g; mean ± S.D. zone diameter 11.72
± 0.461 mm), ceftriaxone (30 g; mean ± S.D. zone diameter 12.44 ± 0.511 mm),
ceftazidime (30 g; mean ± S.D. zone diameter 13.78 ± S.D. 0.428 mm) and cefepime
(30 g; mean ± S.D. zone diameter 11.89 ± 0.676 mm), whilst being sensitive to
ciprofloxacin (5 g; mean ± S.D. zone diameter 23.44 ± 0.511 mm) and meropenem (10
g; mean ± S.D. zone diameter 18.33 ± 18.33 mm) (Fig. 1). All strains were found to
produce ESBL by the double disk synergy test as described previously . A sample
swab from a bucket used to prepare milk was positive for a similar isolate as found in
the blood cultures of the infected neonates. All strains were able to form biofilms on
plastic using the method described by O’Toole et al. .
Most of the neonates (12/17; 71%) were premature infants on artificial feeding (Table 1).
clinical improvement. Owing to the unavailability of carbapenems, the decision was
made to use ciprofloxacin. Despite initiating this treatment, overall mortality was 35%
(6/17). The outbreak was controlled by extensive hygiene measures, including contact
Page 9 of 23
Biotechnology Information (NCBI) GenBank database. A neighbour-joining tree was
precautions, education of health workers, surveillance, thorough disinfection and
washing of all instruments in the unit.
Genetic relatedness of the isolated strains using XbaI-restricted chromosomal DNA
PFGE indicated that all strains, including the one from the milk bucket, were clonal (Fig.
1), indicating the spread of a single strain either from a common source or from one
patient to another. The strains could not be accurately identified using commercial
biochemical kits, with the closest result indicating E. cloacae when using API 20E,
VITEK 2 GN ID Card and Phoenix-NMIC/ID-64, yielding 56%, 95% and 98% identity,
respectively. The isolates could be distinguished from E. cloacae as being non-lactose-
fermenters, esculin hydrolysis-positive and all strains were urease-positive after 48 h of
incubation (Table 2). Fatty acid analysis classified the strain in the Enterobacteriaceae
family with a similarity index of 0.471, 0.438, 0.419, 0.409 and 0.366 to Proteus vulgaris,
Enterobacter gergoviae, Kluyvera ascorbata, Serratia plymuthica and E. coli,
A 16S rDNA phylogeny analysis was performed following sequencing of a 1444-bp
fragment of the 16S rDNA (GenBank accession no. HQ122932) and was compared with
16S rDNA sequences for other Enterobacter spp. in the National Center for
created using MegAlign software (DNASTAR, Madison, WI). The sequence was 97–
98.8% similar to other Enterobacter spp. (Supplementary Fig. 1); the closest similarity
(Table 2) was seen with Enterobacter asburiae (98.86%), Enterobacter cancerogenus
(98.82%), Enterobacter kobei (98.57%), Enterobacter ludwigii (98.55%) and E. cloacae
Page 10 of 23
cause similar diseases and outbreaks as other Enterobacter spp.
(98.52%). A similarity of 99% was observed with the sequence of an uncultured
bacterium (GenBank accession no. EF179826) obtained from the gut of an anopheles
mosquito. Generally, 16S rDNA analysis grouped the strain in the cluster of the E.
cloacae complex. For Enterobacter spp., the general 16S rDNA intraspecific similarity is
ca. 98.7–99% across the entire length of the 16S rRNA gene (ca. 1300–1500 bp)
[12,13]. As 16S rRNA gene sequencing provides genus identification in most cases
(>90%), but less so with regard to species [12,14,15], the rpoB gene, which has been
found to have a high resolution in identifying species in the genus, was used as an
additional marker . A 982-bp fragment (GenBank accession no. HQ148298) was
obtained and was compared with other Enterobacter spp. (Table 2). The closest relation
was E. hormaechei, with a divergence of 1.2%, whilst the 16S rDNA similarity towards
this species was 97.9%, thus making it very unlikely to be the same species, particularly
also because of its different biochemical properties (Table 2) [15,16]. Finally, a 360-bp
fragment of the hsp60 gene was obtained (GenBank accession no. HQ148299) using
previously described primers . The results (Table 2) were 80–97% similar to the
known Enterobacter hsp60 gene sequences .
None of the methods could definitively classify the strains among the known species in
the genus Enterobacter, suggesting that this may be a novel Enterobacter sp. that can
All strains were found to contain blaCTX-M-15 (GenBank accession no. HQ175999), and
PCR analysis with subsequent sequencing could also demonstrate that CTX-M-15 was
associated with an 1800-bp insertion sequence ISEcp1/tnpA gene upstream . In
Page 11 of 23
conjugation experiments, gentamicin, tetracycline and fosfomycin resistances were
transferrable. S1 nuclease PFGE and subsequent DIG hybridisation with a CTX-M-15-
specific gene probe indicated that blaCTX-M-15 was located on a 291-kb plasmid with a
hitherto untypeable replicon in all wild-type isolates. However, in the transconjugants,
blaCTX-M-15 was chromosomally located, thus indicating a translocation of the resistance
gene to the recipient chromosome probably following an abortive conjugal transfer (Fig.
1B) and thereby implying a narrow host range for this plasmid.
Here we describe a new species of Enterobacter that has so far not been categorised
using biochemical and genetic markers. The strain was also isolated from milk and is
capable of biofilm formation; a property also described for other species of Enterobacter
and suggesting a possible association with milk powder. The isolate carries the blaCTX-M-
15 gene on a hitherto untypeable plasmid and is resistant to multiple antibiotics, thus
limiting treatment options. This is the first report of an outbreak of Enterobacter spp.
carrying blaCTX-M-15 in Africa. Surveillance of this strain is needed to determine its
epidemiological spread and to identify markers to improve its detection.
The authors would like to acknowledge the technical support provided by the members
of the Departments of Microbiology/Immunology of Weill Bugando University College of
Health Sciences (WBUCHS) and the Institute of Medical Microbiology (Giessen,
Germany). They also thank Mary Louise Shushu, Claudia Neumann, Hezron Bassu,
Page 12 of 23
Isabell Trur, Kirsten Bommersheim and Alexandra Amend-Foerster for their excellent
This work was supported by a research grant from WBUCHS to SEM, by a researcher
start-up grant from the Faculty of Medicine of the Justus-Liebig-University in Giessen to
CI, and by grants from the Bundesministerium für Bildung und Forschung (BMBF,
Germany) within the framework of the RESET research network (contract no.
01KI1013G) to TC and CI.
Ethical approval was obtained from Bugando Medical Centre/WBUCHS ethics review
Page 13 of 23
 Inversen C, Forsythe SJ. Risk profile of Enterobacter sakazakii, an emergent
pathogen associated with infant milk formula. Trends Food Sci Technol
 Townsend SM, Hurrell E, Caubilla-Barron J, Loc-Carrillo C, Forsythe SJ.
Characterization of an extended-spectrum -lactamase Enterobacter hormaechei
nosocomial outbreak, and other Enterobacter hormaechei misidentified as
Cronobacter (Enterobacter) sakazakii. Microbiology 2008;154:3659–67.
 Neuwirth C, Siebor E, Lopez J, Pechinot A, Kazmierczak A. Outbreak of TEM-24-
producing Enterobacter aerogenes in an Intensive Care Unit and dissemination of
the extended-spectrum -lactamase to other members of the family
Enterobacteriaceae. J Clin Microbiol 1996;34:76–9.
 Hoffmann H, Stindl S, Stumpf A, Mehlen A, Monget D, Heesemann J, et al.
Description of Enterobacter ludwigii sp. nov., novel Enterobacter species of clinical
relevance. Syst Appl Microbiol 2005;28:206–12.
 Ho PL, Shek HLR, Chow KH, Duan RS, Mak GC, Lai EL, et al. Detection and
characterization of extended-spectrum -lactamases among bloodstream isolates of
Enterobacter spp. in Hong Kong, 2000–2002. J Antimicrob Chemother 2005;55:326–
 Morand PC, Billoet A, Rottman M, Sivadon-Tardy V, Eyrolle L, Jeanne L et al.
Specific distribution within the Enterobacter cloacae complex of strains isolated from
infected orthopedic implants. J Clin Microbiol 2009;47:2489–95.
Page 14 of 23 Download full-text
Drancourt M, Bollet C, Carlioz A, Martelin R, Gayral JP, Raoult D. 16S ribosomal
 O’Toole GA, Pratt LA, Watnick PI, Newman DK, Weaver VB, Kolter R. Genetic
approaches to study of biofilms. Methods Enzymol 1999;310:91–109.
 Clinical and Laboratory Standards Institute. Standards for antimicrobial disk
susceptibility tests; approved standard. 9th ed. Document M2-A9. Wayne, PA: CLSI;
 Kiratisin P, Apisarnthanarak A, Laesripa C, Saifon P. Molecular characterization and
epidemiology of extended-spectrum- -lactamase-producing Escherichia coli and
Klebsiella pneumoniae isolate causing health care-associated infection in Thailand,
where the CTX-M family is endemic. Antimicrob Agents Chemother 2008;52:2818–
 Mshana SE, Imirzalioglu C, Hossain H, Hain T, Domann E, Chakraborty T.
Conjugative IncFI plasmids carrying CTX-M-15 among Escherichia coli ESBL
producing isolates at a University hospital in Germany. BMC Infect Dis 2009;9:97.
 Barton BM, Harding GP, Zuccarelli AJ. A general method for detecting and sizing
large plasmids. Anal Biochem 1995;226:235–40.
 Janda JM, Abbott SL.16S rRNA sequencing for bacterial identification in the
diagnostic laboratory: pluses, perils and pitfalls. J Clin Microbiol 2007;45:2761–4.
 Stackebrandt E, Ebers J. Taxonomic parameters revisited: tarnished gold
standards. Microbiol Today 2006;4:152–5.
DNA sequence analysis of a large collection of environmental and clinical
unidentifiable bacterial isolates. J Clin Microbiol 2000;38:3623–30.