Download High-resolution melting analysis of the single nucleotide

Journal of Medical Microbiology (2012), 61, 780–785
DOI 10.1099/jmm.0.041087-0
High-resolution melting analysis of the single
nucleotide polymorphism hot-spot region in the
rpoB gene as an indicator of reduced susceptibility
to rifaximin in Clostridium difficile
Verena Pecavar,1 Marion Blaschitz,1 Peter Hufnagl,1 Josef Zeinzinger,2
Anita Fiedler,1 Franz Allerberger,3 Matthias Maass4 and Alexander Indra3
Correspondence
Alexander Indra
[email protected]
1
Austrian Agency for Health and Food Safety (AGES), Institute for Medical Microbiology and
Hygiene, Waehringerstrasse 25a, A-1090 Vienna, Austria
2
Austrian Agency for Health and Food Safety, Institute for Medical Microbiology and Hygiene,
Spargelfeldstrasse 191, A-1220 Vienna, Austria
3
Paracelsus Medical University, Institute for Medical Microbiology, Hygiene and Infectious Diseases,
Muellner Hauptstrasse 48, A-5020 Salzburg, Austria
4
Institut für Laboratoriumsmedizin, Mikrobiologie und Umwelthygiene, Klinikum Augsburg,
Stenglinstrasse 2, D-86156 Augsburg, Germany
Received 25 November 2011
Accepted 17 February 2012
Clostridium difficile, a Gram-positive, spore-forming, anaerobic bacterium, is the main causative
agent of hospital-acquired diarrhoea worldwide. In addition to metronidazole and vancomycin,
rifaximin, a rifamycin derivative, is a promising antibiotic for the treatment of recurring C. difficile
infections (CDI). However, exposure of C. difficile to this antibiotic has led to the development of
rifaximin-resistance due to point mutations in the b-subunit of the RNA polymerase (rpoB) gene. In
the present study, 348 C. difficile strains with known PCR-ribotypes were investigated for
respective single nucleotide polymorphisms (SNPs) within the proposed rpoB hot-spot region by
using high-resolution melting (HRM) analysis. This method allows the detection of SNPs by
comparing the altered melting behaviour of dsDNA with that of wild-type DNA. Discrimination
between wild-type and mutant strains was enhanced by creating heteroduplexes by mixing
sample DNA with wild-type DNA, leading to characteristic melting curve shapes from samples
containing SNPs in the respective rpoB section. In the present study, we were able to identify 16
different rpoB sequence-types (ST) by sequencing analysis of a 325 bp fragment. The 16 PCR
STs displayed a total of 24 different SNPs. Fifteen of these 24 SNPs were located within the
proposed 151 bp SNP hot-spot region, resulting in 11 different HRM curve profiles (CP). Eleven
SNPs (seven of which were within the proposed hot-spot region) led to amino acid substitutions
associated with reduced susceptibility to rifaximin and 13 SNPs (eight of which were within the
hot-spot region) were synonymous. This investigation clearly demonstrates that HRM analysis of
the proposed SNP hot-spot region in the rpoB gene of C. difficile is a fast and cost-effective
method for the identification of C. difficile samples with reduced susceptibility to rifaximin and
even allows simultaneous SNP subtyping of the respective C. difficile isolates.
INTRODUCTION
Clostridium difficile, a spore-forming, Gram-positive bacterium, is the most frequently identified causative agent of
hospital-acquired diarrhoea and pseudomembranous colitis (Wiström et al., 2001). Numerous C. difficile outbreaks
have been caused by strains of PCR ribotype 027, REA
Abbreviations: CDI, C. difficile infections; CP, curve profiles; HRM,
high-resolution melting; SNP, single nucleotide polymorphisms; ST,
sequence-types.
780
(restriction endonuclease analysis) type BI and NAP1
(North American pulsed-field type 1), as well as toxinotype
III, which is said to lead to more severe disease and higher
mortality rate compared to other C. difficile strains (Kuijper
et al., 2006). The main antibiotics used for treatment of C.
difficile infection (CDI) are metronidazole and vancomycin
(Kuijper et al., 2006). In addition, a promising antibiotic for
the treatment of CDI is rifaximin, a rifamycin derivative.
The mechanism of action of rifamycin is to inhibit the
elongation of 2 or 3 nt RNA transcripts at the 59 end by
Downloaded from www.microbiologyresearch.org by
041087 G 2012 SGM
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
Printed in Great Britain
SNP profiling for rifaximin resistance in C. difficile
binding the RNA polymerase holoenzyme (Campbell et al.,
2001). Recently, four working groups have reported C.
difficile strains with reduced susceptibility to rifaximin due
to SNPs within the encoding gene for the b-subunit of the
RNA polymerase (rpoB) (O’Connor et al., 2008; Curry et al.,
2009; Huang et al., 2010; Huhulescu et al., 2011).
An accurate method for complete genotyping of most
single nucleotide polymorphisms (SNPs) is high-resolution
melting (HRM) analysis (Liew et al., 2004) where DNA
amplification takes place in the presence of a fluorescent
dye that attaches only to dsDNA (Wittwer et al., 2003).
During a subsequent melting step, denaturation of dsDNA
occurs accompanied by reduction of the emitted fluorescence, resulting in characteristic differences measured by
the LightCycler 480 software (Roche Diagnostics).
The aim of the present study was to develop an assay for the
fast identification of SNPs in a proposed hot-spot region of
the rpoB gene, leading to reduced rifaximin susceptibility in
C. difficile by applying HRM.
METHODS
Strains used in this study. A total of 348 C. difficile isolates of
different PCR-ribotypes (Table 1) were obtained from the reference
strain collection of the Austrian national C. difficile reference centre,
which included 25 strains from the ECDC strain collection, 50 strains
from the Leeds University Hospital (UK), two strains from the Public
Health Laboratory Maribor (Slovenia), 242 clinical isolates (129 of
which were previously described by Indra et al., 2008), and 29 strains
from the CDIFFGEN strain collection.
Rifaximin susceptibility testing of the 348 C. difficile samples was
performed in a previous study using a rifaximin broth microdilution
test (Huhulescu et al., 2011). Due to the lack of CLSI or EUCAST MIC
breakpoints for rifaximin against C. difficile we interpreted samples
showing an MIC .32 mg ml21 as having reduced in vitro susceptibility
as suggested by Huhulescu et al. (2011). We investigated the 70 samples
displaying reduced rifaximin susceptibility and the 278 samples that
were rifaximin susceptible by using sequence analysis and HRM
analysis for the detection of SNPs within the rpoB gene.
Capillary gel electrophoresis-based ribotyping of C. difficile
isolates. DNA was extracted from cultures using the MagNA Pure
Compact system (Roche Diagnostics) according to the manufacturer’s
instructions to a final volume of 50 ml. For capillary gel electrophoresisbased PCR ribotyping, primers 16S (59-GTGCGGCTGGATCACCTCCT-39) and 23S (59-CCCTGCACCCTTAATAACTTGACC-39) were
used. The 16S primer was labelled at the 59 end with either carboxyfluorescein (FAM), hexachlorofluorescein (HEX) or tetrachlorofluorescein (TET). Reactions contained 1.5 ml sample DNA, 25 ml
HotStarTaq Master Mix (Qiagen), 0.3 ml (10 pmol ml21) each primer
and 22.9 ml of PCR-grade water. PCR was performed in a standard
thermocycler (Analytic Jena) using the following programme: 95 uC for
15 min, followed by 17 cycles of 95 uC for 1 min, 61 uC for 1 min and
72 uC for 1 min, and a final step at 72 uC for 30 min. PCR products
were sequenced by using an ABI 310 Genetic Analyzer with a 41 cm
capillary loaded with POP4 gel (Applied Biosystems). The samples were
injected at 5 kV over 5 s with a total running time of 30 min at 15 kV
run voltage. As an internal marker for each sample, a 50–625 bp
TAMRA ladder (Chimerx) was used. Data analysis was performed by
using Peak Scanner software 1.0 (Applied Biosystems).
http://jmm.sgmjournals.org
Sequencing of the partial rpoB. All C. difficile samples were
investigated for the occurrence of SNPs within a 325 bp fragment of
the rpoB gene (amino acid positions 458–565) by performing PCRs in
a standard thermocycler and sequence analysis using an ABI 3130
Genetic Analyzer as described previously (Huhulescu et al., 2011).
Briefly, DNA was isolated using the MagNA Pure Compact system
(Roche Diagnostics) according to the manufacturer’s instructions
and eluted to a final volume of 50 ml. The primer pair RifFOR (59CAAGATATGGAAGCTATAAC-39) and RifREVlang (59-GTGATTCTATAAATCCAAATTC-39) was used in PCRs containing 25 ml HotStar
Taq Master Mix (Qiagen), 5 ml (5 pmol ml21) each primer, 13 ml PCR
grade water and 2 ml sample DNA. DNA amplification was performed
in a standard thermocycler using the following programme: 96 uC for
15 min, followed by 30 cycles of 94 uC for 1 min, 52 uC for 1 min and
72 uC for 1 min, with a final step at 72 uC 10 min. PCR products were
purified with exonuclease I and shrimp alkaline phosphatase
(Fermentas) according to the supplier’s instructions.
Sequencing PCRs were performed using a BigDye Terminator v1.1
Cycle Sequencing kit (Applied Biosystems), each reaction containing
2 ml Terminator Ready Reaction Mix, 1 ml sequencing buffer (Applied
Biosystems), 4 ml PCR grade water, 1 ml RifFOR or RifREVlang primer
(10 pmol21) and 2 ml DNA, in a standard thermocycler with the
following programme: 96 uC for 1 min, followed by 30 cycles of 96 uC
for 20 s, 50 uC for 20 s and 60 uC for 4 min. For dye terminator cleanup, PCR products were purified with Centri Sep 96-well plates or
Centri Sep 8 well strips (Applied Biosystems) according to the
manufacturer’s instructions. Samples were analysed in an ABI 3130
Genetic Analyzer (Applied Biosystems) with 36 cm capillary loaded
with POP7 gel (Applied Biosystems).
Data analysis was performed by using Kodon (Applied Maths) version
3.5. Sequences were aligned to the rpoB sequence of a reference wild-type
C. difficile strain (CD630; GenBank accession number NC_009089),
which was obtained from the NCBI database.
HRM analysis of the SNP hot-spot region in rpoB. For HRM
analysis a 151 bp fragment of rpoB (amino acid positions 469–518)
was amplified using the primers RifFORlang (59-GTTAACATAAGACCAGTTTC-39) and RifREV (59-CTCTTTCTCTTGAAAGACC-39).
Each HRM reaction mixture contained 2.4 ml MgCl2 (25 mM), 10 ml
LightCycler 480 HRM Master mix (Roche Diagnostics), 2 ml each
primer (10 pmol ml21) and 0.6 ml PCR-grade water. To generate
artificial heterozygotes, wild-type and sample DNA were mixed in a
ratio of 1 : 6 (0.5 ml wild-type DNA and 2.5 ml sample DNA) and PCR
grade water was used as a negative control.
PCR and melting-curve analysis was performed using the LightCycler
480 System (Roche Diagnostics) with following conditions: preincubation at 95 uC for 10 min, followed by 45 cycles of 95 uC for 10 s, 52 uC
for 20 s and 72 uC for 20 s. After amplification, the melting curve was
produced by heating samples up to 95 uC for 1 min and subsequent
cooling to 40 uC for 1 min. HRM data were obtained at a rate of 25
acquisitions per uC. Data were analysed with the LightCycler 480 gene
scanning software version 1.5 with default settings for sensitivity set to
0.30 and temperature shift set to threshold of 5. The pre-melt normalization ranged from 73.16 to 74.12 and the post-melt normalization
ranged from 81.44 to 82.70. All samples were tested in triplicate and wildtype samples were used as the baseline in difference plots.
RESULTS
Sequence analysis
RpoB sequencing (Table 1) of the 348 C. difficile strains
resulted in the detection of 246 (70.68 %) samples displaying
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
781
V. Pecavar and others
Table 1. Amino acid substitutions in the rpoB gene detected in 348 C. difficile strains based on sequencing results
Amino acid substitutions located in the proposed hot-spot region are written in bold.
PCR ST
HRM CP
RpoB amino acid substitution
Reduced rifaximin
susceptibility
Ribotype (RT) (number of
isolates)
1
2
1
2
Wild-type
R505K
No
Yes
3
3
T501T+L506L+G510G+G512G+F521F+
E541E+K556K
No
4
4
H502N+R505K
Yes
5
1
A555A
No
6
7
5
6
H502L
H502N
Yes
Yes
8
9
7
8
Yes
Yes
10
11
12
13
14
15
16
9
10
11
6
2
1
1
H502Y
S475S+F481F+D492D+T501T+A508A+
G510G+T539T+K556K+S575A
L487F+H502Y
D492N
D492V
H502N+A555A
R505K+I548M
S550F
S550Y
Different RTs (246)
RT027 (46)
RT072 (1)
RT027 (7)
RT126 (4)
RT078 (3)
RT033 (2)
RT078 (2)
RT579 (2)
RT045 (1)
RT053 (1)
RT063 (1)
RT012 (5)
RT027 (4)
RT023 (3)
RT027 (3)
RT115 (1)
RT063 (1)
RT053 (1)
RT027 (1)
RT027 (2)
RT047 (1)
RT027 (2)
RT539 (1)
the wild-type sequence (PCR ST-1) of the rpoB gene. Fifteen
different PCR sequence types (excluding PCR ST-1) were
determined by 23 different SNPs, four of which (L487F,
H502L, S550F and S575A) had not been previously
described (Table 2) and were detected in 102 of the 348
(29.31 %) C. difficile samples. Nine of the 15 PCR sequencetypes had only one point mutation within the 325 bp
fragment of the rpoB gene and six PCR sequence-types
showed additional SNPs (Table 1). The most common PCR
sequence-type was PCR ST-2 which displayed amino acid
substitution R505K detected in 47 of the 102 (46.07 %) rpoBmutated isolates. The second most common sequence-type
was PCR ST-3, containing seven synonymous point mutations
(T501T+L506L+G510G+G512G+F521F+E541E+K556K)
within the 325 bp amplicon of the rpoB gene. This sequencetype was detected in 23 of the 102 (22.54 %) isolates tested.
PCR ST-4, characterized by point mutations H502N+
R505K, and PCR ST-5, characterized by A555A, were both
detected in nine of the 102 (8.82 %) samples tested as shown
782
Yes
Yes
Yes
Yes
Yes
Yes
Yes
RT027
RT027
RT053
RT058
RT027
RT027
RT027
(1)
(1)
(1)
(1)
(1)
(1)
(1)
in Table 1. Sequence analysis revealed point mutations
H502N (PCR ST-7) and H502Y (PCR ST-8) in three
(2.94 %) and two (1.96 %) of the 102 C. difficile isolates
tested, respectively.
Furthermore, PCR ST-6 (H502L), PCR ST-9 (S475S+F
481F+D492D+T501T+A508A+G510G+T539T+K556K
+S575A), PCR ST-10 (L487F+H502Y), PCR ST-11 (D492N),
PCR ST-12 (D492V), PCR ST-13 (H502N+A555A), PCR ST14 (R505K+I548M), PCR ST-15 (S550F) and PCR ST-16
(S550Y) were found infrequently in the rpoB genes of the C.
difficile strains tested. Frequency of occurrence and allocation
to respective ribotypes are shown in Table 1.
HRM analysis
Gene scanning was performed on DNA of 348 C. difficile
isolates (spiked with wild-type DNA for the production of
artificial heteroduplexes) representing 16 different PCR STs
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
Journal of Medical Microbiology 61
SNP profiling for rifaximin resistance in C. difficile
of the rpoB gene (Table 1). After normalization, shifting
and difference plotting of the obtained melting curves, 11
different HRM curve profiles (CP) were detected as
demonstrated in Fig. 1.
Table 2. SNPs detected in the 325 bp rpoB amplicon by HRM
analysis
SNPs located in the proposed hot-spot region are written in bold.
Numbering of nucleotide position starts with the A of the start codon
of the rpoB gene of C. difficile (CD630, NC_009089).
Amino acid
substitution
S475S
F481F
L487F*
D492N
D492V
D492D
T501T
H502N
H502Y
H502L*
R505K
L506L
A508A
G510G
G512G
F521F
T539T
E541E
I548M
S550F*
S550Y
A555A
K556K
S575A*D
SNP
Located in rpoB
amplicon
A1425T
T1443C
A1461C
G1474A
A1475T
T1476C
G1503A
C1504A
C1504T
A1505T
G1514A
T1516C
C1524T
G1530A
A1536G
C1563T
T1617G
G1623A
A1644G
C1649T
C1649A
A1665G
A1668G
T1723G
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
The HRM CP 8, characteristic of PCR sequence-type 9
(S475S+F481F+D492D+T501T+A508A+G510G+T539T+
K556K+S575A), was generated by the presence of six out of
nine SNPs located within the 151 bp HRM amplicon (Tables
1, 2), which displayed a unique melting curve shape in the
difference plot. Furthermore, HRM CP-3 (PCR ST-3), which
showed seven synonymous SNPs (resulting in the following amino acid substitutions: T501T+L506L+G510G+
G512G+F521F+E541E+K556K) within the 325 bp rpoB
fragment in sequencing analysis, produced a clearly distinguishable melting curve in the difference plot based on the
presence of four of the seven SNPs within the shorter HRM
amplicon (Table 2).
Five rpoB sequence types could not be clearly differentiated
by HRM analysis. Sequence-types characterized by a
common CP (HRM CP-1) were: the rpoB wild-type PCR
ST-1 and the three sequence types PCR ST-5 (A555A), PCR
ST-15 (S550F) and PCR ST-16 (S550Y) as demonstrated
in Table 1. Furthermore, PCR ST-13 (H502N+A555A)
showed the same HRM CP (HRM CP-6) as PCR ST-7
(H502N). PCR ST-14 (R505K+I548M) shared a common
HRM CP (HRM CP-2) with PCR ST-2 (R505K).
DISCUSSION
Rifaximin was reported to be a promising drug used for a
follow-up therapy after vancomycin treatment for preventing recurrent CDI (Johnson et al., 2007). Thereby, the rapid
detection of point mutations causing reduced rifaximin
susceptibility in C. difficile strains is of high importance
for the appropriate treatment of CDI and to ensure rapid
therapeutic response. Previous studies showed that amino
*Mutations leading to amino acid substitutions not previously
described for C. difficile.
DS575A is not located within the 325 bp rpoB amplicon.
30
Relative signal difference
25
HRM CP-4 (PCR ST-4)
H502N + R505K
20
HRM CP-6 (PCR ST-7)
H502N
HRM CP-9 (PCR ST-10)
L487F + H502Y
HRM CP-2 (PCR ST-2)
R505K
HRM CP-5 (PCR ST-6)
H502L
HRM CP-7 (PCR ST-8)
H502Y
HRM CP-10 (PCR ST-11)
D492N
HRM CP-11 (PCR ST-12)
D492V
HRM CP-3 (PCR ST-3)
T501T + L506L + G510G +
15
G512G +F521F + E541E
+ K556K
10 HRM CP-8 (PCR ST-9)
S475S + F481F + D492D +
T501T +A508A + G510G +
T539T + K556K
5
0
72
73
74
75
76
77
–5
Temperature C
http://jmm.sgmjournals.org
80
78
79
HRM CP-1 (PCR ST-1)
Wild type
81
82
Fig. 1. Difference plot of 11 HRM curve profile
standards and correlated PCR sequencetypes. Amino acid substitutions located in the
proposed hot-spot region are written in bold.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
783
V. Pecavar and others
Table 3. Frequency of amino acid substitutions in the rpoB gene, previously published and newly described in this study
Amino acid substitution
Frequency
O’Connor et al. (2008)
L487F*+H502Y
S488P
S488T+R505K
D492N*
D492N+R505K
D492V*
S498T+R505K
H502L*
H502N+R505K
H502N
H502N+A555A
H502R
H502Y
R505K
R505K+I548M
S550F*D
S550Y*D
S475S+F481F+
D492D+T501T+
A508A+G510G+
T539TD+K556KD+
S575A*D
Number of strains
Detectable by
HRM analysis
Curry et al. (2009) Miller et al. (2011)
This study
1 (0.28 %)
1 (0.55 %)
1 (1.25 %)
8 (7.84 %)
1 (0.28 %)
1 (1.25 %)
1 (0.28 %)
6 (7.50 %)
1
1
3
1
(1.25 %)
(1.25 %)
(3.75 %)
(1.25 %)
n580
5 (4.90 %)
3 (1.66 %)
8 (7.84 %)
4 (3.92 %)
7 (3.88 %)
2 (1.11 %)
38 (37.25 %)
21 (20.58 %)
6 (3.33 %)
n5102
n5180
1
9
3
1
(0.28 %)
(2.58 %)
(0.86 %)
(0.28 %)
2
47
1
1
1
1
(0.57 %)
(13.50 %)
(0.28 %)
(0.28 %)
(0.28 %)
(0.28 %)
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
Yes
n5348
*Non-synonymous mutations not previously described for C. difficile.
DMutations located outside of the 325 bp rpoB amplicon.
acid substitutions in the rpoB gene, compared with wild-type
sequences, are a reliable indicator of reduced rifaximin
susceptibility in C. difficile (O’Connor et al., 2008; Curry
et al., 2009; Huang et al., 2010; Huhulescu et al., 2011; Miller
et al., 2011). Due to the lack of an official MIC breakpoint for
rifaximin against C. difficile we suggest that samples showing
an MIC .32 mg ml21 are interpreted as reduced in vitro
susceptible (Huhulescu et al., 2011). Based on our results and
previously described amino acid substitutions, we established
an HRM assay using the LightCycler 480 system for the
detection of SNPs within a 151 bp amplicon enclosing the
proposed SNP hot-spot region of the rpoB gene in C. difficile.
Aberrations in HRM CPs of unknown samples, different from
that of the wild-type (HRM CP-1), clearly indicate the
presence of point mutations within the investigated DNA
region, suggesting a reduced rifaximin susceptibility of the
isolates. Comparison of these samples with the 11 different
HRM CP standards defined for this assay provided an
indication of reduced rifaximin susceptibility by the identification of the given sequence types (Fig. 1). During this
study, all samples containing SNPs within the investigated
DNA region were correctly identified as sequences deviated
from wild-type independent from non-synonymous or
synonymous SNPs, these mutations can only be ascertained
by addition of HRM CP standards.
784
Overall, 16 sequence-types were identified in a 325 bp region
of the rpoB gene during this study, 11 of which were also
identified by the described HRM assay (Table 1). The HRM
assay design was based on the location of the most important
point mutations within the rpoB gene (hot-spot region)
(O’Connor et al., 2008; Curry et al., 2009; Huhulescu et al.,
2011; Miller et al., 2011) and the size limitation of the HRM
amplicon length (Liew et al., 2004), leading to the exclusion
of seldom found point mutations located outside of the
HRM amplicon (Table 1). Thereby, two of the 13 STs, PCR
ST-15 and PCR ST-16, were not recognized by the HRM
assay of the present study (Table 1). Nevertheless, the nonsynonymous PCR sequence-type PCR ST-14 (R505K+
I548M) and PCR ST-2 (R505K) shared a common HRM
CP (HRM CP-2) and could be identified, since mutation
I548M was located outside of the HRM amplicon. Interestingly, ST-9 showed reduced rifaximin susceptibility
(Huhulescu et al., 2011) despite displaying only synonymous point mutations in the hot-spot region (Tables 1 and 2);
this remarkable finding could be explained by a nonsynonymous mutation at position S575A in the flanking
region of the investigated rpoB fragment (data not shown).
The HRM assay presented here is suitable to detect all of the
previously published amino acid substitutions (O’Connor
et al., 2008; Curry et al., 2009; Huang et al., 2010; Huhulescu
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
Journal of Medical Microbiology 61
SNP profiling for rifaximin resistance in C. difficile
et al., 2011; Miller et al., 2011), including four new
substitutions (L487F, H502L, S550F and S575A) found
during the course of this study (Table 3). Only the rare
mutation types S550F and S550Y at amino acid position 550,
which were only identified in one sample each during this
study, are located outside of the HRM amplicon.
Interestingly, position 502 seems to be a locus of high
frequency mutation (H502N, H502R and H502Y). With the
amino-acid-substitution H502L we detected an additional
SNP at this locus.
Our data demonstrate that PCR ribotype 027 (n571) is
associated with 11 of 13 detected sequence types (Table 1).
The majority of the 71 isolates (n546) belonged to PCR ST2 (R505K), suggesting that PCR ribotype 027 is predominantly associated with this respective sequence-type. Further
analysis revealed that 42 of these 46 isolates were associated
with a multicentre outbreak in three Viennese hospitals
sharing similar MLVA (multiple locus variable-number
tandem-repeat analysis) patterns (data not shown) (Indra et
al., 2009). These findings underpin the idea that SNP
analysis is a valuable tool for additional epidemiological
investigations (Mutreja et al., 2011). Further studies will be
needed to clarify the correlation between different PCR
ribotypes of C. difficile and the acquisition of reduced
rifaximin susceptibility.
In conclusion, the HRM assay of the present study is a
rapid and specific detection method for the most common
mutations in the SNP hot-spot region of the rpoB gene in
C. difficile and is comparable with sequencing of the
respective rpoB section, which may help eliminate the need
for MIC resistance testing for non-wild-type samples.
ACKNOWLEDGEMENTS
Clostridium difficile strains. Anaerobe 16, 633–635.
Huhulescu, S., Sagel, U., Fiedler, A., Pecavar, V., Blaschitz, M.,
Wewalka, G., Allerberger, F. & Indra, A. (2011). Rifaximin disc
diffusion test for in vitro susceptibility testing of Clostridium difficile. J
Med Microbiol 60, 1206–1212.
Indra, A., Schmid, D., Huhulescu, S., Hell, M., Gattringer, R.,
Hasenberger, P., Fiedler, A., Wewalka, G. & Allerberger, F. (2008).
Characterization of clinical Clostridium difficile isolates by PCR
ribotyping and detection of toxin genes in Austria, 2006–2007. J Med
Microbiol 57, 702–708.
Indra, A., Huhulescu, S., Fiedler, A., Kernbicherler, S., Blaschitz, M.
& Allerberger, F. (2009). Outbreak of Clostridium difficile 027
infection in Vienna, Austria 2008–2009. Euro Surveill 30, pii19186.
Johnson, S., Schriever, C., Galang, M., Kelly, C. P. & Gerding, D. N.
(2007). Interruption of recurrent Clostridium difficile-associated
diarrhea episodes by serial therapy with vancomycin and rifaximin.
Clin Infect Dis 44, 846–848.
Kuijper, E. J., Coignard, B., Tüll, P., ESCMID Study Group for
Clostridium difficile, EU Member States & European Centre for
Disease Prevention and Control (2006). Emergence of Clostridium
difficile-associated disease in North America and Europe. Clin
Microbiol Infect 12 (Suppl. 6), 2–18.
Liew, M., Pryor, R., Palais, R., Meadows, C., Erali, M., Lyon, E. &
Wittwer, C. (2004). Genotyping of single-nucleotide polymorphisms
by high-resolution melting of small amplicons. Clin Chem 50, 1156–
1164.
Miller, M. A., Blanchette, R., Spigaglia, P., Barbanti, F. &
Mastrantonio, P. (2011). Divergent rifamycin susceptibilities of
Clostridium difficile strains in Canada and Italy and predictive
accuracy of rifampin Etest for rifamycin resistance. J Clin Microbiol
49, 4319–4321.
Mutreja, A., Kim, D. W., Thomson, N. R., Connor, T. R., Lee, J. H.,
Kariuki, S., Croucher, N. J., Choi, S. Y., Harris, S. R. & other authors
(2011). Evidence for several waves of global transmission in the
seventh cholera pandemic. Nature 477, 462–465.
This work was supported by the ERA-NET project CDIFFGEN.
O’Connor, J. R., Galang, M. A., Sambol, S. P., Hecht, D. W.,
Vedantam, G., Gerding, D. N. & Johnson, S. (2008). Rifampin and
rifaximin resistance in clinical isolates of Clostridium difficile.
Antimicrob Agents Chemother 52, 2813–2817.
REFERENCES
Campbell, E. A., Korzheva, N., Mustaev, A., Murakami, K., Nair, S.,
Goldfarb, A. & Darst, S. A. (2001). Structural mechanism for
rifampicin inhibition of bacterial RNA polymerase. Cell 104, 901–912.
Curry, S. R., Marsh, J. W., Shutt, K. A., Muto, C. A., O’Leary, M. M.,
Saul, M. I., Pasculle, A. W. & Harrison, L. H. (2009). High frequency of
rifampin resistance identified in an epidemic Clostridium difficile
clone from a large teaching hospital. Clin Infect Dis 48, 425–429.
http://jmm.sgmjournals.org
Huang, H., Weintraub, A., Fang, H., Wu, S., Zhang, Y. & Nord, C. E.
(2010). Antimicrobial susceptibility and heteroresistance in Chinese
Wiström, J., Norrby, S. R., Myhre, E. B., Eriksson, S., Granström, G.,
Lagergren, L., Englund, G., Nord, C. E. & Svenungsson, B. (2001).
Frequency of antibiotic-associated diarrhoea in 2462 antibiotictreated hospitalized patients: a prospective study. J Antimicrob
Chemother 47, 43–50.
Wittwer, C. T., Reed, G. H., Gundry, C. N., Vandersteen, J. G. & Pryor,
R. J. (2003). High-resolution genotyping by amplicon melting analysis
using LCGreen. Clin Chem 49, 853–860.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sun, 30 Apr 2017 01:51:04
785