Академический Документы
Профессиональный Документы
Культура Документы
ORIGINAL ARTICLE
Keywords
environment, identification,
Stenotrophomonas maltophilia.
Correspondence
Sabine Favre-Bonte, UMR CNRS 5557
Ecologie Microbienne, Universite Lyon 1, 43
bd du 11Novembre 1918, F-69622
Villeurbanne Cedex, France.
E-mail: sabine.favre-bonte@univ-lyon1.fr
Abstract
Aims: Aim of the study is to identify accurately Stenotrophomonas maltophilia
isolates recovered from environmental and clinical samples.
Methods and Results: Recovery of Sten. maltophilia-like isolates from soil samples using the vancomycin, imipenem, amphotericin B (VIA) selective agar
medium enabled distinction of various morphotype colonies. A set of soil and
clinical isolates was tested for species identification using different methods.
16S rDNA analyses showed the dark green with a blue halo morphotype to be
typical Sten. maltophilia strains. The API-20NE, Vitek-2 and Biolog phenotypic
analyses typically used for the identification of clinical isolates did not perform
well on these soil isolates. The species-specific PCR screening targeting
Sten. maltophilia 23S rDNA and the multiplex smeD ggpS PCR, differentiating
Sten. maltophilia from Stenotrophomonas rhizophila, were tested for improvement of these identification schemes. The latter multiplex PCR identified all
isolates tested in this study, whatever be their origin.
Conclusions: Isolation on VIA medium and confirmation of Sten. maltophilia
species membership by smeD PCR is proposed to identify environmental and
clinical isolates of Sten. maltophilia.
Significance and Impact of the Study: The proposed approach enables isolation
and identification of Sten. maltophilia from different environments in an easy
and rapid way. This approach will be useful to accurately manage studies on
the abundance and distribution of Sten. maltophilia in hospital and nonhospital
environments.
Introduction
The now known Stenotrophomonas maltophilia species was
originally classified as a member of the genus Pseudomonas
in 1961 (Hugh and Ryschenkow 1961) and then Xanthomonas in 1983, finally coming to rest in Stenotrophomonas
in 1993. The genus Stenotrophomonas belongs to the
c-proteobacteria and includes ten species: Sten. maltophilia
(Palleroni and Bradbury 1993), Stenotrophomonas nitritireducens (Finkmann et al. 2000), Stenotrophomonas acidaminiphila (Assih et al. 2002), Stenotrophomonas rhizophila
(Wolf et al. 2002), Stenotrophomonas koreensis (Yang et al.
2006), Stenotrophomonas humi and Stenotrophomonas
1185
C. Pinot et al.
Sten. maltophilia for the screening of environmental samples. Taxonomic characterization of Sten. maltophilia isolates is mostly performed using a biochemical approach
(i.e.: Bollet et al. 1995; Denton and Kerr 1998; Qureshi
et al. 2005). Sometimes a molecular approach based on
the 23S rRNA gene amplification (Whitby et al. 2000)
completes the species identification scheme. This speciesspecific PCR (SS-PCR) approach was developed to overcome the problems associated with definitive identification. This PCR presented sensitivity and specificity of
100% for clinical strains or strains isolated from the environmental clinical area (Whitby et al. 2000; Giordano
et al. 2006; Foster et al. 2008a). However, there is a lack
of studies describing isolation and identification of
Sten. maltophilia isolated from environmental samples.
In the present paper, we reported on the use of the
VIA medium and 16S rDNA sequencing to select and
identify Sten. maltophilia-like isolates from soil samples.
We further applied various biochemical methods classically used in a clinical context and molecular-based
approaches reported from the literature to evaluate their
identification efficiencies. The accuracy of these methods
was also verified on a set of clinical isolates.
Materials and methods
Experimental sites
Soils were sampled from different locations in various areas
in France including agricultural sites in Burgundy and in
Ile de France, experimental waste water irrigation sites in
France and two sites in Tunisia; and agricultural sites
amended with organic matter in Burkina Faso. A total of
104 soil samples were collected from the upper layer (05,
010 or 020 cm), sieved (2 mm), stored at room temperature and analysed within 2 weeks. They were collected
during various campaigns between 2005 and 2008. Sites
and sampling strategy are briefly described in Table 1.
Isolation of Sten. maltophilia-like isolates from soil
samples
One hundred and four different soil samples were tested
for isolation of Sten. maltophilia-like isolates. Briefly, soil
samples (5 g) were blended with 50 ml of a saline solution (NaCl 08%) for 90 s in a Waring Blender (Eberbach
Corporation, Ann Arbor, MI, USA). Homogenized soil
suspensions were serially diluted in sterile saline solution
and dilutions were spread on VIA plates (Kerr et al.
1996) supplemented with cycloheximide (200 mg l)1) to
inhibit fungi development. Plates were incubated 2448 h
at 28C. Four morphotypes were obtained on VIA agar
plates including fermenting bacteria producing yellow or
C. Pinot et al.
Sites
Land
use
Clay Silt
(%) (%)
Sand pH
Number of
(%) water samples
France
Feucherolles
Pierrelaye
Balot
Venarey les laumes
Is sur Tille
Bourberain
Dompierre
en Morvan
Ruffey les Echirey
Marcilly Ogny
Commarin
Echevronne
Morey St Denis
St Aubin
Enesad
Wheat
Maize
Barley
Grassland
Rape
Bulk soil
Grassland
19
8
29
44
44
39
18
75
13
69
52
48
58
46
6
79
2
4
8
3
36
69
82
58
68
81
7
56
15
21
1
1
1
1
1
Rape
Triticale
Bulk soil
Cereal
Vineyard
Vineyard
Bulk soil
47
36
42
45
33
35
35
39
48
50
41
43
42
36
14
16
8
14
24
23
29
81
79
71
79
83
82
8
1
1
1
1
1
1
1
Burkina Faso
Tabtenga
Toudoubweogo
Zagtouli
Yagma NZ
Yagma Eapit
Bulk soil 11
Bulk soil
9
Sorghum nd
Sorghum
9
Sorghum 12
165
20
nd
30
18
693
71
nd
61
70
61
75
77
74
72
6
6
6
6
2
Tunisia
Souhil
Beni Khiar
Nabeul
Citrus
Orange
trees
Orange
trees
12
8
8
9
80
73
82
84
10
10
14
77
67
10
1187
C. Pinot et al.
16S-rRNA
API-20NE
Vitek-2
Biolog Gram-negative
Multiplex PCR
BPOE 5100
BPOE 5101
BPOE 5102
BPOE 5103
BPOE 5104
BPOE 5105
BPOE 5106
BPOE 5107
BPOE 5108
BPOE 5109
BPOE 5110
BPOE 5111
BPOE 5112
BPOE 5113
BPOE 5114
BPOE 5115
BPOE 5116
BPOE 5117
BPOE 5118
BPOE 5119
BPOE 5120
BPOE 5121
BPOE 5122
Clinical
Stenotrophomonas maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Stenotrophomonas rhizophila
Sten. rhizophila
Sten. rhizophila
Sten. rhizophila
Sten. rhizophila
Sten. rhizophila
Variovorax sp.
Dyella japonica
Sten. maltophilia
Burkholderia cepacia
B. cepacia
Pseudomonas luteola
Ps. luteola
Ps. luteola
B. cepacia
Ps. luteola
Ps. luteola
Ps. luteola
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Ps. luteola
Sten. maltophilia
Ps. luteola
Ps. luteola
Sten. maltophilia
Ps. luteola
Sten. maltophilia
Sten. maltophilia
Ps. luteola
Sten. maltophilia
Sten. maltophilia
Pseudomonas aeruginosa
Sten. maltophilia
Ps. fluorescens*
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Brevundimonas diminuta
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Ps. fluorescens
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
B. diminuta*
no id
Aeromonas salmonicida
Sten. maltophilia
Pseudomonas fluorescens
Sten. maltophilia
Pseudomonas syringae
Sten. maltophilia
Ps. fluorescens
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Sten. maltophilia
Flavobacterium tirrenicum
Pseudomonas synxantha
Sten. maltophilia
Ps. fluorescens
Sten. maltophilia
Sten. maltophilia
Pseudomonas
Sten. maltophilia
Sten. maltophilia
Ps. synxantha
Sten. maltophilia
Sten. maltophilia
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
smeD+
ggpS+
ggpS+
ggpS+
ggpS+
ggpS+
ggpS+
ggpS+
smeD+
smeD+
Biochemical characterization
API 20NE
The API test kit (bioMerieux) consists of enzymatic and
carbon compound assimilation test elements (20 tests).
For each isolate, inoculation of API 20NE strips was
performed as recommended by the manufacturer (bioMerieux). After incubation for 48 h at 28C, results were
read and analysed using the apilab plus software
package. Only confidence levels expressed as very good
identification or excellent identification were considered
in this study as described in Zbinden et al. (2007).
Vitek analysis
Identification was performed using ID-GNB Vitek2 cards
(bioMerieux). The Vitek-2 technology (bioMerieux) is
based on API technology but comprised 64 tests and is
automated. It leads to an identification result in 36 h for
Gram-negative (GN) bacteria. The Vitek cards were filled
according to manufacturers instructions for testing in the
Vitek2-compact automate (bioMerieux). Results were
described with a high confidence level (excellent identification or very good identification) except when
mentioned.
1188
Biolog
The Biolog Omnilog Identification System (AES Chemunex,
Bruz, France) utilizes automated biochemical tests. It is a
walk-away instrument that tests the ability of bacteria to
use a panel of 95 carbon sources. Growth is detected by
using a tetrazolium violet salt which is incorporated into
each well of a 96-wells Omnilog microtiter plate. Reduction in tetrazolium violet gives a purple colour indicating
the presence of growing bacteria. Prior inoculation of the
Omnilog plates, each isolate was cultured at 28 or 37C
on a Biolog Universal Growth agar plate with 5% sheep
blood. After an incubation period of 1824 h, a bacterial
colony was suspended at the specified density using Omnilog inoculating fluid as described by the manufacturer.
The suspension was subsequently used to inoculate wells
of GN2 microplates (AES Chemunex). According to their
origin, the microplates were incubated at 28 or 37C.
Each metabolic profile was compared with the appropriate GN Omnilog Biolog database which contains
biochemical fingerprints of 526 GN species. Biolog identifications were considered if the similarity index of the
genus or species was 0750 or greater after an incubation
of the plates for 4 h. When a lower similarity value was
C. Pinot et al.
Clinical
isolates
Identification system
Sm (15)
Sm (28)
API-20NE
Vitek-2 compact
Biolog Gram-negative plates
smeD PCR
5
11
10
15
28
28
27
28
(33)
(73)
(67)
(100)
(100)
(100)
(96)
(100)
1189
C. Pinot et al.
C. Pinot et al.
selective media, VIA has the advantage of allowing a differentiation of Sten. maltophilia isolates from nonSten. maltophilia isolates which is often not the case for
selective media. On cetrimide-nalidixic acid agar,
Ps. aeruginosa can exhibit different morphotypes (Lavenir
et al. 2008) but on the LAM medium, the different
Acinetobacter species and B. cepacia complex isolates can
show similar ones (Jawad et al. 1994). It is noteworthy
that a high selectivity of a culture medium can sometimes
be a disadvantage as it could lead to an under-estimation
of bacterial diversity as evidenced in the study of Tarnawski et al. (2003) who compared the diversity of fluorescent Pseudomonas populations recovered from two
different selective media.
We were interested in identifying species exhibiting
other morphotypes on VIA than the Sten. maltophilia
dark green colonies with a blue halo to better define the
extent of the diversity of the species able to grow on that
medium. Three of these morphotypes were found to
belong to D. japonica, Sten. rhizophila and Variovorax sp.
by 16S rDNA sequence analysis. However, Sten. terrae,
Sten. humi, Sten. acidaminiphila or Sten. nitritireducens,
which were previously isolated from soils e. g. Heylen
et al. (2007), were not recovered from our soils. This was
not surprising because neither white bacterial colonies
with a blue halo nor light green colonies with a blue halo,
typical of these species growing on VIA, were detected
during our platings.
Screenings on VIA medium and analyses of 16S rDNA
sequences appeared reliable approaches for identifying
Sten. maltophilia. However, such a strategy is not usually
applied for routine identification especially in hospital
laboratories which prefer high throughput, less timeconsuming and inexpensive methods. Biochemical
approaches fit these requirements and can be easily
automated. Sensitivity analyses of these approaches on a
selection of 23 environmental isolates obtained from soil
samples and representative of each morphotype were performed. API20NE was the least efficient method, with only
33% of Sten. maltophilia isolates being properly identified.
API-20NE sometimes allocated the isolates to B. cepacia
and Ps. luteola. Not least, some Sten. maltophilia morphotypes closely related to Sten. rhizophila or D. japonica,
were allocated to the Xanthomonadaceae family. Previous
studies reported Sten. maltophilia isolation on VIA
medium followed by the identification with API-20NE
from environmental samples including faucets, domestic
sink drains or salads (Denton et al. 1998; Qureshi et al.
2005) but no description of the morphotypes was available
to relate these data with our observations. The accuracy of
the automated Vitek-2 system was also checked because it
involves a higher number of biochemical tests than
API20NE. This system was found more efficient than
1191
C. Pinot et al.
1192
C. Pinot et al.
1193